ONYX builds ONYX Ai Matte as a GPU-accelerated matte extraction tool for segmentation, tracking, and mask refinement inside OFX-compatible hosts. It targets hair and fur edge work and shots with more than one subject, and it frames the speed and quality claims as marketing claims. Under the hood, it uses Meta SAM3 and VitMatte models, packaged via ONNX in FP16 and accelerated with NVIDIA CUDA 12.6 and TensorRT 10.x.

Prompts, trimaps, and tracking

The UI exposes five pipeline modes: box or point prompts with auto track, box or point prompts with manual track, text prompts with auto track, mask prompts with auto track, and a trimap workflow using VitMatte with manual track. Point and box workflows let you set foreground and background points and a bounding box. The interface supports up to 32 points.

Text prompt mode detects objects that match your text. If it finds multiple matches, one foreground point picks the instance. Mask prompt mode takes a rough input mask and uses it as the prompt for a refined segmentation. The trimap path expects white foreground, black background, and a grey uncertainty band for edge refinement. Tracking integrates directly into the prompt workflows. In Nuke, the box center can be driven by expressions so a tracker can push it through the shot while size stays consistent. You still need to test any new tool before it goes into production, especially when the alhpa becomes the single point of failure.

Hosts and compatibility

The published host list includes DaVinci Resolve 18 and later, Fusion Studio 18 and later, and Nuke 13 and later. It also positions the plug-in as working in any OFX compatible app, like Autodesk Flame specifically.

System requirements and what is not supported

Minimum requirements list Windows 10 64 bit, an NVIDIA RTX GPU, 6GB of VRAM, 16GB of RAM, and 8GB free space. Windows 11 64 bit is also supported. Text prompts require 8GB of VRAM or more. The baseline GPU callout is an NVIDIA RTX 2060 or better with SM 7.0 or higher. Not supported: GTX 1000 series, AMD Radeon, and Intel Arc.

Trial, licensing, and pricing

A full featured 7-day trial is offered with no credit card required. Licenses are sold as lifetime, one time payments with no subscription, plus free minor updates. Pricing is published in euros as lifetime tiers on the official pricing page. Indie costs €80 for one machine node locked. Studio 3 costs €180 for three machines node locked. Studio 10 costs €450 for ten machines node locked. Studio 10 also includes priority support.

https://onyxofx.com/

The post ONYX Ai Matte for OFX hosts first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool, SynthEyes from Boris FX is a standalone 3D camera-tracking and matchmove application used in VFX pipelines. It exports to major DCC and compositing tools and sits upstream of layout, animation, and comp, quietly solving cameras so others can focus on lighting.

Boris FX has announced SynthEyes 2026 with a faster, more responsive camera solve release aimed at reducing iteration cycles and workflow interruptions in feature film and high end streaming projects. The update introduces a Mocha Point Tracker, Live Preview for geometric hierarchy solving, user interface revisions and expanded USD support. According to the company, the release is shaped by user feedback and focuses on daily production tasks rather than broad architectural changes.

Faster supervised tracking

As shots become more complex, trackers require more manual attention. Rapid camera moves, whip pans and motion blur increase the likelihood of drift or failure. But a classical point tracker operates by defining a small target area and a larger search area. On each frame, the tracker searches for the target area’s pixel pattern within the search region by scanning left, right, up, and down. This works reliably when the camera is locked off or moves in a controlled manner on a dolly, gimbal or tripod.

The limitation is that a standard RGB point tracker does not inherently account for rotation or scale changes in the pattern over time. These are affine transformations: translation, rotation, scaling, and shear. If the tracked feature rotates or changes size significantly, the tracker may drift or lose the pattern entirely.

SynthEyes provides options such as updating the reference frame at defined intervals. However, between those reference updates, the tracker still searches primarily by translation, allowing drift to accumulate. In shots with strong rotation or scale change, this can result in unstable supervised tracks and broken solves.

For these scenarios, SynthEyes 2026 introduces the Mocha Point Tracker. The feature leverages Mocha’s planar tracking technology under the hood directly into the supervised tracking workflow. The goal is to combine the robustness of planar tracking with the point data required for camera solving.

Planar brain, point output

Planar trackers are designed to compensate for rotation, scale and other affine transformations across a defined surface patch. Instead of following a single pixel pattern, they analyse the motion of a planar region.

The Mocha Point Tracker tracks a small planar patch and returns the coordinates of the centre of that patch as point tracking data. In practical terms, artists still obtain a single 2D point suitable for solving, but that point is derived from a planar solve rather than an RGB comparison.

In the user interface, switching a tracker from the traditional RGB mode to Mocha Point Tracker updates the tracker controls without launching Mocha. The planar patch can be reshaped freely. Corners can be adjusted individually, and the centre point can be offset by holding Shift while dragging.

The planar patch does not need to remain a strict rectangle. Artists can enlarge it to capture more contrast detail, for example, around the edges of an architectural feature. During tracking, the patch visibly rotates and scales with the image content, while the returned centre point remains tightly locked.

This behaviour addresses a core limitation of classical point trackers: Because the planar patch models translation, rotation, scale and affine deformation, it remains stable under perspective and motion changes that would cause a standard tracker to drift.

Parameters for unruly shots

The Mocha Point Tracker includes adjustable parameters for more demanding material. By default, each tracker automatically determines how much of the initial pattern must be visible in subsequent frames before continuing. This percentage threshold can be manually adjusted.

Users can also select different motion models. The default model is 2D TRSA, which stands for Translation, Rotation, Scale, and Affine. Alternative models constrain the tracked motion to fewer parameters. For heavy perspective change, a 3D Perspective model is available.

All models ultimately return single-point data, but they use different internal motion assumptions to generate that data. In addition, advanced controls allow users to increase the expected per-frame rotation and scale change. By default, SynthEyes expects up to one degree of rotation and half a per cent scale change per frame. These limits can be raised for more aggressive camera movement. These advanced controls are typically unnecessary, as the planar tracking approach is quite forgiving at default values, but who knows what happened on set? Sometimes it is best not to ask.

Mixing tracker types

SynthEyes continues to support mixing different tracker types within a single supervised tracking setup. The Mocha Point Tracker is particularly useful for bridging gaps where feature trackers fail. In such cases, adding larger planar patches can help maintain continuity across problematic frames, enabling a continuous solve where feature tracking alone breaks down.

This highlights a workflow caveat: When combining two Mocha Point Trackers into a single tracker, SynthEyes first converts them into classical point trackers before merging. This avoids merging two planar patches directly, which would be unstable. After conversion, the resulting track retains point data but no longer allows modification of the original planar patch.

Boris FX positions the Mocha Point Tracker as a complementary tool rather than a replacement. Traditional RGB trackers remain better suited for manual frame by frame adjustments, where artists may need to babysit a track through occlusions. The supervised tracking system is designed to allow mixing and matching tracker types as needed.

In short, when the camera behaves politely, classical point tracking remains efficient. When it does not, the Mocha Point Tracker offers a planar fallback inside the same workflow, without requiring artists to leave SynthEyes. If neither works, demand reshoots.

Real time GeoH feedback

SynthEyes 2026 introduces Live Preview for geometric hierarchy, referred to as “GeoH”. Live Preview is a toggleable checkbox located above the axis lock controls in the GeoH interface. When enabled, the GeoH setup dynamically adjusts while users manipulate trackers and selected parameters.

With Live Preview active, artists can adjust trackers, focal length and distortion parameters and see the hierarchy update instantly in the current frame. The primary use case described by Boris FX is tracker adjustment. Instead of replaying the shot after every modification, the result is visible immediately.

In addition, the feature operates hierarchically. Users could typically enable Live Preview on the object level where downstream elements should animate in real time. For example, activating it on a shoulder joint would allow the arm chain beneath it to update in real time. In practice, it can also be enabled on the topmost root object if required.

In current builds, enabling Live Preview highlights the affected hierarchy in yellow, providing visual feedback about which nodes are active. This is particularly useful in larger hierarchies where dependencies can become difficult to track.

But why?

The system updates without requiring repeated play-stop-check cycles. Users can move trackers, adjust seed positions, tweak pivots, unlock or rebalance axes and modify lens or offset parameters while seeing the result immediately. You can think of GeoH in SynthEyes as a lightweight rigging system inside the tracking environment. Objects are created at mesh pivots and parented hierarchically. Animation channels must be explicitly unlocked before they can be driven by trackers or mesh deformation. Locked parameters remain static.

A typical workflow involves creating a GeoH object from a mesh, branching joints from parent nodes, adjusting pivots in Edit Pivots mode and painting vertex weights to define influence regions. Trackers can then be parented under specific GeoH objects, with seed points projected onto the mesh to drive animation.

With Live Preview enabled on a GeoH object, moving a tracker updates the mesh deformation or joint rotation immediately in the viewport. The effect is comparable to interactive rig manipulation in DCC applications, with the mesh responding as trackers are repositioned.

Before finalising animation for export, users should disable Live Preview on all relevant nodes and then play through the shot so that the animation is fully realised. This step ensures the resulting keyframes are baked from the tracked motion. Or, don’t do that, and we’ll see what happens :) Jokes aside, the feature should cut iteration time by at least one order of magnitude.

SynthEyes Special: The Interface refresh begins

The 2026 release marks the beginning of a “multi-phase user interface modernisation”. The current update focuses on the left panel and room navigation – one thing after the other. Boris states that improved labelling, more consistent layout and interaction behaviour, and a clearer visual structure are intended to enhance discoverability and daily usability. At the same time, existing workflows and muscle memory are said to be preserved.

The changes are positioned as incremental rather than disruptive. Advanced users are expected to retain familiar workflows, while first-time users may find the application more approachable. Because, as good as Syntheyes is, the Interface was always a bit … particular.

USD moves upstream

And another thing: SynthEyes 2026 expands USD support, allowing USD assets and scenes in the application. Users can import individual meshes or full USD scenes while preserving hierarchy, cameras, lights, animation and metadata, including parent-child relationships and transforms. USDZ assets import with embedded textures intact.

Whole scene transforms and STMaps

A new Whole Scene Transform mode allows scale, rotation or translation to be applied to the entire scene, including cameras, objects, trackers and meshes, while preserving relative motion and solve integrity.

This mode is intended for late-stage adjustments, including multi-camera projects. The company states that relative motion and solve integrity remain intact, though no technical detail is provided on the underlying implementation.

Distortion map export, commonly referred to as STMap export, is now integrated directly into the standard Export menu and the Multi Export system. This enables distortion data to be generated alongside other outputs in a single export pass rather than as a separate step. STMaps are commonly used in compositing to apply or remove lens distortion via UV maps. Integrating STMap export into the main export pipeline reduces the number of discrete export operations required.

Pricing and availability

SynthEyes remains a standalone application. Subscription, perpetual, upgrade and support renewal options are available. Pricing starts at 325 USD per year or 62 USD per month. Customers with an active SynthEyes or Boris FX Suite subscription, or with an upgrade and support plan, receive the 2026 release as a complimentary update.

Further details are available via the SynthEyes product page on the Boris FX website.

As with any update that affects solving, hierarchy behaviour or export fidelity, new tools and innovations should be tested before use in production.

// SynthEyes 2026 What’s New
// https://borisfx.com/products/syntheyes/whats-new/

The post SynthEyes 2026 Speeds Up first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool: The Pixel Farm develops PFTrack, a node-based 3D camera tracking and photogrammetry system used in VFX, virtual production, and survey-driven pipelines. It exports to DCC tools including Autodesk Maya and integrates LiDAR and image based reconstruction.

The Pixel Farm has released PFTrack 26.02, a point update to its camera tracking and photogrammetry software. The release focuses on export reliability, Survey Solver behaviour and data validation rather than new solver technology. According to the published change list, the update introduces improvements to Maya export scripts, OBJ texture handling, LiDAR import logging and node connectivity within the Survey Solver. It also resolves a keyboard shortcut conflict in the Orient Camera node.

Maya export scripts corrected

The updated Maya export scripts now automatically parent background image planes to cameras on export. This addresses a long standing friction point in matchmove handoff. Previously, depending on scene configuration and export path, artists could end up manually re-parenting image planes inside Maya to maintain alignment between the solved camera and its background plate. The revised script automates this relationship at export time.

For tracking teams delivering camera solves into Maya-centric pipelines, this reduces post-export clean-up and the risk of misalignment during layout or scene assembly. The vendor statement does not detail changes to other DCC export scripts.

Survey Solver and OBJ texture loading

PFTrack 26.02 updates the way OBJ mesh textures with numbered filenames are loaded into the Survey Solver node. OBJ sequences or assets that rely on incrementally numbered texture files will now load correctly when imported into Survey Solver. The official notes do not specify naming conventions, but refer to numbered filenames in general terms.

For facilities using photogrammetry meshes or survey derived assets with multiple texture tiles, correct automatic loading reduces manual reassignment inside the node graph. It also limits the possibility of mismatched textures during camera solve refinement. The change appears limited to texture loading behaviour inside Survey Solver and does not reference modifications to the OBJ importer elsewhere in the application.

LiDAR import logging added

The update adds explicit error logging when importing corrupted or incomplete LiDAR survey data files. PFTrack supports LiDAR survey data within its camera-solving and scene-reconstruction workflows. In previous versions, incomplete or damaged files could fail during import without detailed diagnostic output.

With 26.02, the system now logs errors when such data is detected. The vendor documentation does not describe the logging format, verbosity level or where logs are written. It does state that error logging is triggered when corrupted or incomplete LiDAR files are imported.

Photogrammetry node connectivity fixed

A further fix addresses connector logic in Studio and Enterprise editions. Photogrammetry nodes can now be connected to secondary inputs of the Survey Solver where appropriate. The official wording specifies that this applies to Studio and Enterprise only.

This suggests that earlier builds restricted valid node graph connections in certain configurations, potentially limiting more complex photogrammetry driven solves. The updated connector logic restores the ability to route photogrammetry outputs into secondary Survey Solver inputs when supported by the node’s design.

The vendor has not published technical detail about which specific input sockets were affected or how the internal validation rules have changed. Not independently verified at press time.

Facilities using advanced node graphs that combine survey data, image based reconstruction and manual constraints should review existing templates to confirm expected behaviour after updating.

Scope and positioning

This release does not introduce new solver models, GPU-acceleration changes, or major UI revisions. It is positioned as a maintenance and workflow refinement update. The focus remains on export reliability, data validation and node graph correctness. For studios that rely on PFTrack for camera tracking, survey alignment and photogrammetry reconstruction, these adjustments address specific operational irritations rather than expanding feature breadth.

PFTrack is available in three editions. PFTrack Solo is priced at £699 for a permanent licence. PFTrack Studio is offered on a rental basis, with pricing listed at £59 for a five day rental period. PFTrack Enterprise pricing is available on request from The Pixel Farm. As always, facilities should confirm current licensing terms and support options directly with their vendor before procurement.

As always, new tools and updates should be tested in controlled conditions before being introduced into active production pipelines.

The post PFTrack update tidies Maya and LiDAR workflows first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

DaVinci Resolve (DR for short) had the Magic Mask function on board for quite a while, and some voiced their disappointment when it didn’t conjure up what greenscreen keying does. Well, Blackmagic Design (BM for short) never claimed it does, by concept it was rather aiming at secondary colour grading. While its capabilities are quite impressive, Magic Mask has some problems with backgrounds that are too close in colour and contrast to the foreground, or subjects with motion blur. And then, it needs very capable hardware and there are still quite a few reports of limited reliability of Magic Mask 2 in DR, caching in particular. While version 20.3.2 should have some improvements in that area, we wouldn’t bet the house (or the company) on it yet.

FxFactory to the rescue

So, good old greenscreen has not yet been obsoleted by AI, at least if you don’t shoot the whole scene in a virtual studio anyway. The authors, Rob Mackintosh and Gabriele de Simone, have obviously thought about facilitating boring tasks by blending in some AI with ‘classic’ greenscreening. They are the masterminds behind the Hawaiki Keyer (HK for short, even if not from China), which is supported and distributed by FxFactory. Version 5 of that keyer adds tracking by a neural engine, including face or object recognition. They also ported it to DR, while older versions were only available for Adobe’s video software and Final Cut Pro. Reason enough to have a closer look?

Test scenes

We have tested it with critical footage, like a too well-lit background lacking saturation, and another one with good saturation, but lighting issues. The latter one was even in ‘small’ HD out of a DVCPro camera (from Hollywood Camera Works). Of course, a perfectly lit greenscreen in 12K from a modern digital camera would hardly stress any of the keyers included by BM, like the 3D Keyer in DR Studio, or Cryptomatte and Delta Keyer in Fusion. HK rather aims at those fighting with not so perfect shots.

Tracking instead of boring

After watching the older tutorial you may be wondering about the shape of the window coming up for the secondary keyer, which is now rectangular. You can round it off if needed with its “Roundness” slider. This is now in its own menu under “Shape”, where you also define the tracking area. The secondary key is controlled by this adjustable shape, and this is where tracking comes into play. You can have it follow the initial content like any good tracker.

Or you can activate “AI Track” and let it follow faces or objects, even multiple targets up to six. In our tests it was tracking very reliably, only some of the wildly shaken strains of blonde hair with high motion blur in one example were overshooting and needed minor adjustment. Such tracking capabilities are not only taking a lot of manual work out of garbage matting for scenes with a moving camera, but allow you to apply the secondary key to critical areas, like hair. A detailed tutorial is here.

Two for the price of one

Actually, even four, counting the extras. At first view, most of HK’s functions are set to “Auto” or quite decent presets, but all can be fine-tuned manually, if not to say infinitely. There is even a separate version for bluescreen, in case anybody, not even knowing where the heck that name originated, is confronted with optochemical footage. Now, either version is actually two keyers in one. If you have some experience with chroma keying, you will already know that one single key is often not enough to handle both delicate semi-transparencies and an uneven background at once. HK offers a primary and a secondary key, which both can be enabled and fine-tuned individually, as explained here. Most of the YouTube tutorials for HK were made by Simon Ubsdell, a very experienced compositing artist who also produced hundreds of other tutorials.

Clean the edges

“Denoise” is not exactly what you may assume. General denoising, if needed, should rather be done before using your incoming footage with HK. We would suggest NeatVideo for such tasks, which is still king in the lower price range. That function in HK is rather aiming at critical edges from weaker sources, like 4:2:2 subsampling, which can lead to stairstepping or even crawly contours. While it can’t perfectly cure our poor sample from DVCPro, it did a pretty good job improving it, considerably better than simply blurring the edges. There is a whole set of settings to improve the edges under “Edge Tool”, you can find that part of a tutorial by Simon here.

Clean your screen

And then there is “Screen Clean”, explained here. It does a better job than the standard sliders of other keyers at identifying and cleaning areas of uneven lighting of the background and preserving semi-transparent foreground areas at the same time. Screen Clean has a pretty decent “Auto” default, but can be fine-tuned and excluded from an active shape if needed. Of course, there’s “Despill”, which will not only base its effect on the keying color, but can accomodate the background image if you wish.

Grading included

HK even integrates a whole range of color grading functions. This may have you wondering why, if you are working inside one of the best grading applications on the planet. But anybody with some decent compositing experience will know how helpful it can be to do some corrections right in the keyer. Of course, you can also choose appropriate color space and gamma here. Most of these parameters were added in version 5, BTW. Like all the rest, those values can be animated.

If you have never used this keyer before, have a look at the tutorials for HK 4 and 4.5, they contain valuable explanations of the earlier improvements. Not all material is yet adapted to DR, but this tutorial is going over the use of the keyer in DR in general. In particular the use of the background for functions like the “Edge Tools”, e. g. light wrap. These can be a bit tricky in Resolve and will be easier with the help of a simple Fusion composition, clearly explained in that video.

Help is just a mouse-click away for all of the Hawaiki tools, both in the Inspector or the Open FX Overlay. The manual will show up in a separate window, which you can place on a second screen or next to DR on a superwide one, if DR is not set to full-frame.

DR with Fusion not enough?

Now, if seasoned Fusion specialists may claim that they could do most of what HK does with the tools already on board, that may be true to some degree. But that’s not easy for the average users of DR, who will probably see HK’s all-in-one concept as a godsend. They can discover a highly-specialised set of tools coming clearly from practical experience and not found anywhere else. And then, HK works in the free version of DR too, while many of the more advanced tools in DR don’t.

Being fully optimized for Apple silicon, the keyer is very fast, even on a humble Mac mini M4. The intelligent pipeline contributes to this: if you first crop the picture in for garbage matting, the outside areas are not processed any further by the tracker and all the rest to avoid the load. And then, whatever can be done by the neural engine will not put any load on the GPU cores. Finally, on limited hardware in particular it’s considerably more stable than Magic Mask 2.

Commentary

There might be only one minor flaw to disappoint a few of our readers: all FxFactory filters are MacOS only, with no plans to change that any time soon. But for all those using DaVinci Resolve, even the free version, on a modern Apple computer and needing greenscreening regularly, Hawaiki Keyer’s all-in-one approach should be quite appealing. Many of the other offers by FxFactory are worth a look as well.

The post Hawaiki Keyer – for those who don’t trust all the magic of AI first appeared on DIGITAL PRODUCTION and was written by Uli Plank.

For those who don’t know the tool: 3DEqualizer by Science-D-Visions is a high-end 3D tracking and matchmoving tool used in VFX and compositing workflows. It connects to Maya, Blender, Nuke and others to generate precise camera and object solves for integration of CG elements.

Over the winter break, Science-D-Visions released 3DEqualizer4 Release 8.1. The update focuses on precision and pipeline reliability, with several technical improvements aimed at matchmove and postvis specialists.

Piggyback rigs and postvis tools

The standuot addition is a new control pane for computing main camera motion from a piggyback or auxiliary camera, including 360-degree setups. This workflow allows the calibration of static camera-to-camera transformations directly from calibration or production shots. Science-D-Visions also provides a demonstration video of this workflow. To simplify onset capture, a new import script lets users load footage directly from movie files, reducing dependency on external transcoding tools.

Small but targeted refinements

Release 8.1 introduces a new high-pass filter in image controls, a frequently requested addition for analysing fine detail in tracking plates. It also adds start-frame parameters to USD and Alembic exports, allowing more precise control when aligning sequences in downstream departments. A new Attribute Editor plugin demonstrates how to reference 3D models directly from USD files, intended as an example for studios developing their own custom pipeline tools. The update further includes a rewritten Blender export script (v2.0) that matches the design of the Python-based Maya exporter from Release 8.0.

Technical updates and Python 3.11

Under the hood, 3DEqualizer 8.1 moves to Python 3.11. The runtime update brings better session management, new methods for converting OpenCV-style lens distortion to native 3DE formats, and improved image buffer handling. Several older tools were refined, including decimation of 3D models with correct UV handling, enhanced curve filtering, and better distance constraint behaviour. The development team also addressed multiple stability issues: bug fixes cover 2D track exports to Flame, curve editor display errors, long-lat conversion crashes, and problems with rolling shutter exports.

Export and scripting improvements

Maya export now includes user preferences, the “object pgroup z distance” tool gained motion path transfer functionality, and file requesters now display file sizes in human-readable format. The ScriptDB installer received fixes for SSL warnings, and the “reel in points” and “import 2D tracks” tools were corrected for minor issues.

Pricing and compatibility

3DEqualizer 8.1 runs on Windows 10+, glibc 2.17+ Linux, and macOS 13.5+. Perpetual licences cost €7 699 (around USD 9 065), while subscriptions range from €65 per week to €1 704 per year. As always, new builds should be tested in non-critical environments before being rolled out to production shots, especially given the shift to Python 3.11 and updated exporters, which may affect pipeline compatibility.

// https://www.3dequalizer.com/index.php?id=251111_01&site=devblog

The post 3DEqualizer 8.1 Adds Piggyback Camera Solves first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool: SynthEyes by Boris FX is a dedicated 3D camera tracking and matchmoving software used across film, VFX, and advertising production. It interfaces smoothly with Nuke, After Effects, Fusion and other DCC applications, delivering fast, precise solves for integration of CG into live action. Part of the Boris FX Suite, SynthEyes sits alongside tools such as Mocha Pro, Continuum, Sapphire, and Silhouette, forming a tightly integrated pipeline of visual effects and finishing plugins.

Free matchmove mastery – for a limited time

Boris FX has announced SynthEyes Essentials, a new training series designed to teach both the conceptual foundations and practical workflows of matchmoving in SynthEyes. The package, usually priced at USD 95, is available as a free premium download until 31 December 2025.

The offer includes over three hours of professional instruction, spread across eighteen videos, with accompanying project files and sample footage. The series is taught by Boris FX product specialist Christoph Zapletal, a long-time VFX artist and trainer – and Digital Production’s own Compositing, Flame and Star trek Author.

Topics include matchmove theory, camera solving, AI-assisted workflows, roto masking, lens distortion management, mesh generation, and object tracking. Boris FX positions the series as a comprehensive, concept-driven training rather than a feature-by-feature tutorial. If you want to learn what you are doing, and not just which button to press, this is the course for you.

What do you get?

Across eighteen lessons and twelve blocks, SynthEyes Essentials builds a structured learning path from theory to application. It starts with conceptual understanding, progresses through tracking, solving, calibration, and ends with practical export and integration workflows. Each block is designed to be modular, allowing artists to focus on specific parts of the matchmove process or follow the series sequentially for a full onboarding to SynthEyes.

Foundations and Core Concepts

The course opens with an Introduction and a clear explanation of the Underlying Concepts of matchmoving. These first lessons establish the theoretical foundation of what camera solving actually does, namely mapping 2D image data into a consistent 3D coordinate space. Christoph outlines how SynthEyes interprets motion, focal length, and parallax, giving artists a solid base before they begin working hands-on. This section sets the tone for the entire training: conceptual understanding first, tools second.

Tracking Workflows

Once the principles are clear, the series moves into practical tracking. Lessons on Automated Tracking and Supervised Tracking demonstrate SynthEyes’ two primary approaches to extracting tracking data. Automated tracking covers point generation and solving for shots that can be processed algorithmically. The supervised section focuses on when and how to step in manually: tuning track points, correcting drift, and controlling motion paths when the automated system struggles. Together, these chapters establish a complete understanding of tracking strategies for real production footage.

Solving and Refinement

The next stage is “The Solver Room”, where all the accumulated tracking data is converted into a working 3D solve. The solver section demonstrates how SynthEyes interprets the camera’s path and reconstructs the scene’s geometry. The training then introduces Roto Masking, showing how to isolate moving foreground elements or exclude unwanted regions from the tracking process. This not only improves accuracy but also keeps solves stable in complex or occluded shots.

Lens and Calibration

Zapletal then focuses on optical parameters in three consecutive chapters: Lens Distortion, Lens Calculation, and Lens Calibration. These lessons explain how real-world lenses introduce distortion, how SynthEyes models and compensates for it, and how to calculate or calibrate lenses from footage or reference data. This block is particularly useful for artists working with varying cameras or mixed-format source material, where lens data is inconsistent or missing.

Coordinate Systems and Scene Setup

The two lessons titled Coordinate Systems Part 1 and Part 2 explain how SynthEyes handles scene orientation, scale, and alignment. Understanding coordinate spaces is crucial when exporting solves to downstream 3D applications such as Nuke or Maya. These chapters ensure that solved data lines up correctly with CG scenes and compositing environments, avoiding scale and rotation mismatches in later stages of production.

Mesh Generation

The Mesh Generation Part 1 and 2 chapters explore SynthEyes’ built-in mesh tools. Artists learn how to create proxy geometry that helps verify solves, anchor objects in space, and visualise tracking quality. These lessons demonstrate both automated and manual mesh workflows, preparing users for typical use cases like ground plane setup or background modelling for compositing reference.

Object Tracking

Next, Object Tracking Part 1 and Part 2 move beyond camera solves to show how SynthEyes handles independent object motion. This section is key for shots with multiple moving elements—cars, props, or handheld devices. It explains how to isolate and track objects independently while maintaining a consistent world space between multiple solves.

Export and Integration

The final three chapters deal with transferring the finished solve to other applications. Zapletal demonstrates exporting cameras, point clouds, meshes, and nulls to standard 3D and compositing packages. Topics include export format choices, scaling conventions, and verification of data integrity. The course closes with best practices for integrating SynthEyes results into downstream VFX or editing pipelines.

Availability and next steps

Artists can download SynthEyes Essentials for free via the Boris FX webshop until the end of December 2025. From 1 January 2026, it will be priced at USD 95 for the premium download, though the videos will remain freely viewable on the Boris FX website. As with any training-based workflow adoption, users are advised to validate techniques and settings against their own production data before deploying in critical pipeline stages, but, after we have seen a few of the videos, this one will be useful for pretty much everybody.

The post SynthEyes Essentials Training is here and Free for the Holidays first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool: KeenTools is a suite of professional tracking and head-creation plugins used in Blender and Nuke for geometry and facial workflows, and in Adobe After Effects for object tracking. It complements DCCs like Blender for 3D production and Nuke for compositing, and interacts with character-design tools such as Character Creator.

What’s in 2025.3

Keentools released version 2025.3. This is the final scheduled release for the year and focuses on host compatibility and workflow extensions. The update adds official support for Blender 5.0 across its Blender plugin pack. Earlier versions of KeenTools plugins already supported Blender 2.80+, but this brings full compatibility with the latest major Blender release.

FaceBuilder for Blender now includes direct integration with Character Creator 5, allowing artists to send 3D head models created in Blender into Character Creator 5. This expands on existing integrations and streamlines head-creation pipelines between these tools.

After Effects & Export Improvements

GeoTracker for Adobe After Effects gains support for After Effects 25.6, ensuring the plugin runs with the current AE build. The notable addition is the ability to export 3D scene-tracking data as an Alembic (.abc) file, making it easier to transfer tracked objects and camera data into other 3D applications. The release notes emphasise that Alembic export applies to full 3D scene tracking from GeoTracker in After Effects.

Nuke Open Beta Support

For Foundry Nuke users, the 2025.3 update adds compatibility with Nuke 16.1 and 17.0 open beta builds. This ensures KeenTools nodes and plugins operate in these pre-release environments as users test the latest compositor versions.

What’s Not Changed

There is no indication from the official release notes of major new tracking algorithms, UI overhauls, or fundamental workflow rewrites. This update centres on host support and select workflow exports rather than algorithmic feature expansion.

What You Should Know

Artists considering production adoption should verify compatibility with their exact host versions and pipeline tools before integrating 2025.3 into active projects. As always with plugin updates, running controlled tests in a non-production environment is recommended. Not independently verified at press time: performance impact or stability across all host configurations.

The post KeenTools 2025.3 Brings Blender 5, Character Creator 5 & AE Exports first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

Mocha Pro 2026 is out, and Boris FX’s Emmy and Academy Award-winning planar tracker returns with an update that focuses on refinement rather than reinvention. The 2026 release adds new AI-driven roto and matte tools, smarter 3D-solve cleanup, and a rebuilt Curve Editor, all tuned to make shot-fixing and cleanup faster and more predictable for working compositors.

The update also aligns with VFX Reference Platform 2025, adding support for Qt 6.5.4 with PySide 6, Python 3.11, OpenEXR 3.4.5, Alembic 1.8.8, and FBX 2020.3.7. This ensures fewer pipeline compatibility issues in studio environments.

Matte Refine ML: AI for tricky edges

The new Matte Refine ML system expands Mocha Pro’s machine-learning masking family, which already includes Object Brush ML, Matte Assist ML, and Face ML. It analyses existing masks and automatically refines soft or semi-transparent edges which is useful for hair, fur, motion blur, and defocus.

Two refinement modes are available: one optimised for soft motion or blur, the other for sharper edges. Additional post-processing sliders (Black Clip, White Clip, Edge Grow, Shrink/Grow, and Blur) let artists fine-tune how transparency and detail are blended along the edge.

For multi-layer composites, the new Group Layer Mattes option merges multiple layers into a single matte clip, improving render and playback performance. Combined with Matte Assist ML, this creates an efficient path from object masking to refined alpha without intermediate renders.

Refine Solve: faster, cleaner 3D tracking

The Refine Solve tool targets 3D camera solves that have drifted or accumulated bad data. Using the Clean Up Features dialog, users can delete unstable or short-lived tracking points, then recalibrate the existing solve without starting over.

Because Refine Solve builds on the original solve data, it preserves previous adjustments such as ground-plane alignment or scene scaling. For extended tracking and finishing, the refined data can be exported directly into SynthEyes for further cleanup. The result: faster iterations and fewer re-solves for 3D match-move tasks in complex shots.

Blips and 3D

Technically, Mocha’s camera solver identifies trackable “blips” in the image and converts them into 2D trackers, which are then triangulated into static 3D feature points. The software measures the average deviation between the 2D trackers and 3D points as HPix (horizontal pixel error). A low HPix value signals a stable camera solve, while spikes in the error curve expose moments where the 3D reconstruction drifts. These spikes often correspond to unstable track data caused by motion blur, occlusion, or minor parallax inconsistencies.

The Clean Up Features system provides targeted filters to address these issues. It can automatically disable feature points during high-error frames, discard very short or inconsistent trackers, and remove features that contribute excessive deviation. Once unwanted data is removed, Refine recalibrates the existing camera without re-solving from scratch, updating only the relevant calculations while maintaining previously established orientation and coordinate systems. This allows technical directors to preserve scene alignment and object placement while improving solve precision.

Refinement should be applied only when the base camera movement is physically plausible. If the initial camera path is incorrect or erratic, a full re-solve remains necessary. However, for typical production shots where the solve is broadly correct but slightly unstable, this incremental workflow offers a reliable way to tighten accuracy, lower HPix variance, and maintain alignment integrity across multiple refinement passes.

Curve Editor rebuilt for modern pipelines

After a long hiatus, Mocha Pro’s Curve Editor makes a full return. The interface has been rewritten from the ground up, with tighter integration between the Dope Sheet and Graphs panel. Artists can now visualise tracking, roto, and camera-solve data as curves, then smooth, zoom, or isolate problem areas directly. A new offset-curve overlay in the Adjust Track module displays how refinements affect the original track in real time. Filtering options allow users to display only keyframed parameters or selected spline points, helping to pinpoint problematic data without clutter.

Smaller but useful refinements

Beyond the headline features, Boris FX lists numerous “quality-of-life” updates under the hood. Playback of refined mattes no longer requires separate rendering, and composite mattes can now be applied directly in other Mocha render modules, such as Insert or Remove.

Mocha Pro 2026: Pricing and availability

Mocha Pro 2026 is available as a standalone application and as plugins for Adobe After Effects, Premiere Pro, Avid Media Composer, Foundry Nuke, Blackmagic Fusion, Autodesk Flame, and VEGAS Pro. Subscription pricing starts at €42 per month or €290 per year. Perpetual licences (from €675), upgrades, and support plans are also offered. Customers with active Mocha Pro or Boris FX Suite subscriptions receive the 2026 update at no additional cost.

Reality check

While the AI-driven refinements and rebuilt Curve Editor mark significant usability improvements, studios should validate performance and matte accuracy in controlled conditions before integrating Mocha Pro 2026 into production pipelines.

The post Mocha Pro 2026: Refined, re-solved, re-edged first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

The VFX toolkit SLAPSHOT, developed by Hotspring, has added a new module: an AI-powered camera tracking tool designed for 2D compositing workflows. According to Hotspring, it can generate production-ready camera solves and point clouds in minutes, a process that traditionally consumes many hours using manual or semi-automated tracking methods.

From zero data to full solve

The new Camera Tracking module targets artists who primarily work in 2D compositing but occasionally require 3D camera data to integrate digital elements into live-action plates. Hotspring says the tool can produce accurate camera tracks without needing any prior camera metadata such as lens information or focal length. Users can, however, supply camera data if they have it, and the software will incorporate that information into the final solve.

Output formats include .abc and .glb camera files, .exr STMaps for warping and undistortion tasks, undistorted .jpg plates, and a .mov playblast. This ensures compatibility with most standard 3D and compositing environments. Hotspring states that the tool was built to handle the bulk of real-world VFX work, background replacements, screen inserts, sky swaps, and clean-ups, where precision matters but speed is often the main constraint. The system is hosted entirely in SLAPSHOT’s cloud environment on AWS, supporting resolutions up to 8K and removing the need for heavy local compute resources.

Why this might matter

Camera tracking has long been one of the most time-consuming stages in VFX production. Solving for camera movement, lens distortion and point clouds often requires manual supervision and multiple iterations. Hotspring claims that SLAPSHOT can complete many of these tasks in under ten minutes, compressing what was once a day-long operation into a near-real-time process.

For smaller studios or freelance compositors who lack dedicated matchmove teams, this could represent a measurable reduction in turnaround time and cost. However, the company makes clear that the system is not designed to replace full-featured matchmove pipelines for heavy CG integration.

Not for every shot

Jon Mason, co-founder and CEO of Hotspring, emphasises that SLAPSHOT’s AI Camera Tracking is aimed at medium-difficulty shots, not full 3D sequences with extensive camera-lens complexity. According to Mason, the tool’s purpose is to close the gap between 2D tracks and complex CG matchmoves, where automation can safely accelerate work without sacrificing usable accuracy. Mason also noted that building the system required combining multiple open-source AI components to achieve consistent and dimensionally accurate results. There was, he said, “no off-the-shelf way to do something like this.”

Pricing

SLAPSHOT operates on a subscription model with four pricing tiers. The Free plan allows artists to test the tools at no cost, capped at 2K resolution and 1 GB of storage. The Lite plan, aimed at entry-level users, costs $9 per month and raises the limit to 3K resolution with 5 GB of storage.

The Starter plan, recommended for freelancers and solo compositors, is priced at $39 per month, allowing up to 4K output and 50 GB of storage. The Pro plan, intended for professional users and studios, costs $129 per month and supports 8K resolution, 250 GB of storage, and unlimited seats. Each plan includes a defined number of export frames, after which additional renders are billed at $0.02 per frame.

Part of a broader toolkit

Camera Tracking joins SLAPSHOT’s existing suite of AI-assisted post-production tools, which includes AI Rotoscoping for matte generation, Generative Retiming for extending footage by synthesising new frames, and Video Depth Maps for stable per-pixel depth estimation. Together, these cloud-based tools form a platform for accelerating everyday compositing and finishing work without high hardware costs. All of SLAPSHOT’s modules run on Hotspring’s independently audited AWS infrastructure, which is designed for commercial, film, and episodic-scale productions. And sometimes goes down, as we have seen last week. So far no information of a on-prem option.

Hotspring’s background

Hotspring is best known as a global VFX outsourcing platform used by studios worldwide. Its founders — Jon Mason, Varun Parange, and Ben Stallard, have backgrounds in delivering Oscar-, Emmy-, and VES-winning visual effects. Drawing on this experience, they built SLAPSHOT to automate repetitive VFX tasks while keeping final quality in artists’ hands. The company says that Camera Tracking continues its mission to “tackle the real-world problems artists face every day” by turning complex, technical workflows into faster, accessible processes.

What professionals should test

For compositing supervisors and VFX artists, SLAPSHOT’s new tool could be worth evaluating in production conditions. The key questions will concern accuracy under difficult circumstances, such as handheld or fast-motion footage, extreme lens distortion, or heavy motion blur, and how well exported data fits into existing workflows across Nuke, After Effects, Blender, and other software. Pipeline integration, output fidelity, and network performance in high-resolution workflows remain areas that users will need to test directly before adopting the system for ongoing projects.

Unverified details at press time

Some technical specifics remain unconfirmed from official sources, like full benchmark results across shot types, and the level of pipeline integration offered beyond the stated file formats. These details were not publicly available on SLAPSHOT’s or Hotspring’s websites at the time of publication. Therefor, as always, artists should run extensive tests before integrating any automated tracking system into mission-critical workflows.

The post SLAPSHOT promises AI Camera Tracking “in minutes” first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

Silhouette has always been marketed as the tool for people who can’t escape roto. With Silhouette 2025.5, Boris FX has put machine learning helpers front and centre: object detection, face segmentation, and prompt-based masking join the node tree. The goal is simple: fewer clicks, more consistent mattes, and less time spent duplicating hand masks across 400 shots of the same jawline.

Silhouette 2025.5: Prompt it, mask it, propagate it

The Mask ML node now accepts natural language prompts. Type “car” or “tree” and Silhouette produces a mask. Type a longer description, and the AI will attempt a semantic search to find matching pixels.

Selected elements are added to an object list, which then feeds into Matte Assist ML. Instead of baking a single alpha, objects can be output as Cryptomatte data, letting artists toggle them on and off directly in Silhouette or export them for compositing in any other app with Cryptomatte support. This system is not just UI-driven: prompt-based masking is also scriptable, making it viable for automated pipelines. In theory, you could send hundreds of shots through a command-line batch, prompt them for “car,” and get consistent object mattes without opening the GUI. In practice, test before you trust it.

Faces in pieces

The new Face ML node detects and segments eyes, lips, mouth, and surrounding skin into separate mattes. Like Mask ML, it can output standard alpha channels or Cryptomatte data. This is meant for invisible beauty work: subtle retouches, clean-ups, and feature-by-feature adjustments without manual roto. The node delivers tracked mattes, meaning the eyelid follows the eye across frames without an assistant spending the weekend keyframing bezier handles.

SynthEyes inside Silhouette 2025.5

Perhaps the most significant addition: 3D camera solving powered by SynthEyes, now built directly into the 3D Scene node. Until now, Silhouette users imported camera solves from external trackers. With 2025.5, clicking “solve” runs a SynthEyes-based tracker on the footage, returning camera data, point clouds, and per-frame point tracks.

Once solved, users can create 3D cards from selected points, assign colours or labels, and then use unprojection/reprojection to paint or composite directly into stabilised views. The solved camera allows for matchmove paint strokes: work on the unprojected plate, then reapply strokes in the moving scene.

The system reports solve accuracy as average error in horizontal pixels. As a rule, anything under one pixel is usable. For example, an average error of 0.7165 px across 400+ frames is considered solid. The node also includes solver presets: nodal pans, locked shots, or full free-move.

To avoid bad data, Silhouette introduces an occlusion matte input. Feeding a mask into the node tells the tracker which areas to ignore (sea, sky, or other irrelevant regions). This can dramatically lower error values and concentrate tracking points on useful geometry. In short: no more hopping between Silhouette and external trackers for simple 3D cards and matchmove paint.

Nodes refreshed

Several existing nodes have been upgraded:

• Cryptomatte: Supports overlays on input images and selective matte activation directly in the viewer.
• Tracker: Gains an occlusion matte input and direct 3D integration, allowing the Point Tracker to generate a tracked layer from selected 3D points.
• Mask ML: Now outputs multiple objects via Cryptomatte, plus text-prompt detection.
• Matte Assist ML: Introduces compute caching for smooth playback without recomputation, and separate Cryptomatte outputs when multiple objects are present.

Machine learning gets faster

Three major AI nodes received new models and performance updates:

• Matte Assist ML: Object count raised from 16 to 48.
• Denoise ML: Adds “Better” and “Faster” v3.0 models optimised for compression+noise single images.
• UpRes ML: Gains NVIDIA RTX Video Super Resolution support on Windows. Updated v2.0 models ship in “Better” and “Faster” variants. All options process frame by frame for consistent results.

Combined with a new ML node caching system, playback of AI-driven mattes is significantly smoother in both standalone Silhouette and its plug-in form.

Workflows and scripting

Silhouette 2025.5 quietly upgrades scripting support. Prompt-based masking and AI mattes can be triggered via command line, enabling automated pipelines for large projects. For production facilities with centralised render farms, this makes AI masking a candidate for batch pre-processing before shots reach compositors.

Silhouette’s place in the toolbox

Silhouette remains Boris FX’s specialised roto, paint, and compositing package, the software you call when planar trackers and bezier masks stop being funny. Boris FX itself has been busy. Alongside Silhouette, the company maintains Mocha, Continuum, and Sapphire, a trinity of plug-ins already considered staples in post. With recent moves into audio tools as well, Boris FX is positioning itself as an unavoidable extension layer for almost every VFX and finishing pipeline.

The post Silhouette 2025.5: Roto With Prompts, Faces, and a Built-In SynthEyes first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

We had the pleasure of talking to Simon Hall from Blackmagic Design about Fusion, AI and the community. Simon Hall knows post-production from both sides of the suite. Before stepping into his current role in business development at Blackmagic Design—where he focuses on post-production and supports the Resolve ecosystem across EMEA—he spent over a decade getting his hands dirty in the edit bay.

Starting as an offline editor and later working as a Smoke operator in a northern UK studio, Simon eventually moved to London to take up a training role—merging his technical know-how with a knack for sharing knowledge. His past as a freelance editor kept him close to the tools, even as his focus shifted to product strategy. Over the past ten years, he’s been a key figure in shaping DaVinci Resolve’s growth from a specialist grading tool to a full-fledged post-production platform with a massive user base.

DP: Fusion 20 marks a huge step forward, and the community has been very enthusiastic. What was the motivation behind this release, and what can users look forward to in future updates?

Simon Hall: We are a company that really does value user feedback. What the devs always want to do is to take user ideas and integrate into the software. As you can imagine, those lists tend to be very long. SO we have to pick and choose what we do. Our motivation as a company is to give people the features and workflows they want. We will continue to try to get the best out of our software. Because we like making people happy at the end of the day.

DP: The Integration in Resolve was a great idea – as it was to keep the Standalone Edition alive which is favoured by many professionals for performance reasons. Can we expect full feature parity (i.e. all Resolve OFX plugins available in the standalone Fusion)?

Simon Hall: Some of the open effects are obviously designed for colour. So that’s why you may not find them in fusion studio. But the devs try and get parity as much as they can. Over time, what will happen is hopefully we’ll start to get parity between the two applications. The open effects that are available in resolve are then also available in fusion.

DP: When you’re developing both editions, is there an ideal use case in your mind for when users should use each version?

Simon Hall: If you’re doing things like shot clean-ups, you’re doing some matte painting, you’re doing some tracking, my recommended workflow is to use fusion in resolve.
I always tend to steer people towards fusion studio when you’re doing sort of heavy VFX work – You’ve got a couple of hundred nodes, you’ve got 3d models, you’ve got particle generators. As soon as you start to do something that becomes quite complex, you’re going to get performance enhancements in fusion studio. The one thing that fusion studio can do that resolve can’t, is to use network rendering.

DP: Both Resolve and Fusion include very useful ML tools like the Magic Mask. What’s next on your roadmap for neural-network features? Any change to see an integration with tools like ComfyUI? What is Blackmagic Design’s long-term vision for applying machine learning across the industry?

I could talk about what we’re thinking, but the truth is-we don’t really know. There’s no set roadmap. That’s just kind of how things work at Blackmagic. We don’t always know what’s coming. Our approach to AI is focused on speeding up workflows-not replacing the person in the chair. That’s the goal behind all the AI we’ve built so far. Also at the moment all of our AI processing happens locally-on your system.

We don’t send your files anywhere to train the AI.

There is one exception, and that’s the scene extension feature. It’s not in the public beta yet, partly because it uses Blackmagic Cloud. Since the system needs to understand the image, it does have to send it up to the cloud for processing. We’re still figuring out how that will work, especially in terms of using it with a cloud account, which may involve some cost. Our AI is meant to help-not take over. So when it comes to the kind of AI that generates full scenes or content, like the comfy UI stuff, I could be wrong, but I don’t see us going in that direction.

DP: The community loves that Blackmagic hasn’t adopted a subscription model—but new features still cost money to develop. Are there any plans to introduce an upgrade-pricing scheme in future releases?

Simon Hall: Potentially in the future, we might charge for upgrades. A lot of this software development takes time, money, and people-we’ve got a growing team of developers. But I don’t think this is going to turn into a subscription. Grant himself really doesn’t like the idea of locking people out of their own work; he thinks it’s unfair. Long-term, yeah, we want to make things sustainable, but nothing’s been decided yet.

It kind of became a hot topic at NAB because Grant mentioned it in his video. What surprised us was how many people actually supported it-people who really want Resolve and Fusion to keep developing and improving. And if that means maybe paying, say, a $50 update once a year, or per version, a lot of folks were like, “Yeah, you know what? I’d be fine with that.” Especially considering we’ve never really charged for updates before.

DP: You surely know the Fusion Forum WSL/Steak underwater. Do you incorporate feedback from the user community into your ongoing development?
Simon Hall: Yes, absolutely. And yeah, great name-“We Suck Less.” The Fusion community there is full of really smart people writing macros, scripts, all sorts of things. A lot of our product specialists are on that forum. I don’t know if you know Steve Roberts-he was one of the original guys behind Eyeon. When we acquired Fusion from Eyeon, Steve came over with it, since it was basically his baby. He’s still a member of the forum and still checks in. We all do, really. I’ve even picked up scripts from there.

There’s one I use all the time-it animates to the beat of music by analysing the audio. I use it constantly, though I’ve forgotten what it’s called. So yeah, we’re definitely across that forum. And when people post feedback or feature requests there, the developers do take a look-just like we do across other forums-to see what the community is talking about and what they’re asking for.

DP: Is there one feature you’d most like to add or change for yourself?
Simon Hall: Yeah, one thing I’d really like to see in Fusion 20 is a bit more “smart assistance” when building node trees-especially in 3D scenes. Sometimes I’ll go to connect something, like a camera, and it just doesn’t work. Then I realize I’ve missed a node-maybe I forgot to add a 3D plane or something in between.

What I’d love is if Fusion could recognize that kind of situation. Like, if I try to connect two nodes, and it doesn’t work, the system could say, “You’re probably trying to do this,” and automatically drop in the missing node-like a 3D merge or a plane-so the connection makes sense.

Basically, a bit of smart logic that fills in the gaps when I miss something. That would really help, especially when working quickly. So yeah, that would be my feature request-something like a smart assistant that helps you build the flow correctly when it sees what you’re trying to do.

DP: Do you have a favourite movie (or specific shot) that was created using Fusion?

Simon Hall: Oh, there are quite a few-especially if we’re talking old-school Fusion. One of my favourite shots was in the film Swordfish. It’s about hacking, with a young Hugh Jackman and John Travolta. The opening shot, where someone’s left a claymore mine and the camera moves through the explosion in what looks like a frozen moment-that whole sequence was done in Fusion. That was back in the Eyeon Fusion days, and I always thought it was a fascinating use of the tool.

In more recent films, there are a few standouts. The Martian used Fusion heavily, and Top Gun: Maverick-which has been one of my favourite films in the last few years-used it for a lot of shot cleanups. The grading on that was done in Resolve. There’s also a TV show called Bosch on Amazon Prime. It’s been one of my favourite series for a while. Fusion was used extensively on that as well-for scene extensions, keying, and shot clean-up. So yeah, those are some of the highlights for me.

DP: Thanks for your time, Simon :)

The post The (further) development of Fusion. first appeared on DIGITAL PRODUCTION and was written by Manuel Kotulla.

We first take a close look at the new functions and then test various workflows in general. As a special treat we have an interview with Simon Hall from Blackmagic Design about the development of Fusion – which you can read very soon. Stay tuned! But now, let’s start up Fusion and see what’s what.

Native support for Cryptomatte

Or the “compositor’s lifesaver” – with a cryptomatte, complex masks can be created quickly and easily from all objects or materials present in a 3D scene. The only prerequisite for this is that Cryptomattes are included when the 3D scene is rendered. Manual object selection or ID assignment is not necessary.

The Cryptonode itself presents these masks as extremely colourful surfaces that can be selected with a simple click and converted into an alpha channel. The mattes can be combined as required and removed again from a simple list view. Complex adjustments such as the colour grading of objects in the blur can be implemented quickly with cryptomattes, although they can also reach their limits here (we will look at a workaround for this special case later in the chapter on multilayers).

USD & 3D compositing

Let’s stick with this example. Suppose we want to make the light a little more dramatic or add particle FX or text in the room or or or or … Here, in addition to the possibility of masking things to infinity, the fantastic (real) 3D space of Fusion comes into play, which can be fed with all kinds of data from FBX to USD. Let’s take a look at the updated USD workflow.

The USD file generated by Houdini Solaris is simply dragged and dropped into Fusion and is then loaded and ready for use. Incidentally, the 3D space opens automatically as soon as a 3D object is to be displayed on one of the viewers (shortcut “1” or “2”).

To make the rear lion’s head a little brighter, we can now add a USD light and illuminate the head as desired. As we are working with real 3D data, the light is modulated correctly, which is a huge advantage over working with conventional masks. To avoid potentially disturbing influences on other objects, these are faded out using uVisibility. A newly added 3D layer acts as a light blocker to allow the light to flow gently downwards. Finally, this black and white image generated by a uRender Node is added directly over our original rendering – or better, used as a mask for a colour corrector.

3D text can be added in a similarly simple way, and parts of Fusion’s proprietary 3D system can be integrated into the USD Scene – as is the case with a Text3D Node. Particularly nice: If a Z-channel is also output via uRender, Fusion can perform a so-called depth merge, in which the merge node automatically “masks” the objects in the correct order using the Z-channel.

To round off the trip, we export the whole fun as USD and load the file back into Houdini as a sublayer and can access the new light and the Lightblocker here.

What we would like to see in a future update: A USD stage manager and more Hydra delegates aka render engines that support Fusion.

An alternative to relighting via 3D is the relatively new relight node, which uses a rendered or calculated normal map. This is practical if you don’t have a 3D scene to hand and can do without precision.

(Real)Deep Compositing

Deep compositing is a brand-new feature in Fusion for intervening deeply in the image – and completely without 3D. Here, the pixels are also given depth values during rendering so that Fusion now knows where these pixels are located in 3D space.
The decisive advantages are mask- and Z-channel-free merging of objects and the creation of real depth position-based masks (the above-mentioned depth merge accesses a Z-channel, a 2D representation of the depth in space), which, in contrast to the volume mask (which accesses the world position), are very precise and virtually free of edge problems. Volumes such as smoke and explosions can also be handled wonderfully, as objects can interact correctly with volumes. The new freedom comes at the price of larger data volumes and necessarily a render engine that supports deep data (e.g. Karma, Vray, Arnold, Octane, Renderman and Redshift).

But right from the start – how do we even know that we are dealing with deep rendering? Fusion writes “deep” + colour depth after the resolution in the viewer and shows a cheerful “Z” in the channel selection (not multilayer selection).

The new node “DeeptoPoints” then shows the whole truth – as a point cloud. We do not see “haptic” 3D objects, but the distribution of individual image pixels in space.

Let’s assume we now want to mount a fireball behind the lion heads. In “normal” compositing, volumes (explosions, smoke …) are a guarantee for alpha channel problems – not so in deep compositing.

Above: The delivered explosive DeepEXR rendering (via Karma) and the Deep2Points view – the explosion has an actual volume of points, which will be very useful to us in a moment. The two deep renderings are merged using dMerge Node – and automatically know which elements should be in the foreground thanks to the 3D depth information they contain.

The fireball can even be moved in depth using dTransform and is automatically covered by the corresponding objects – or even partially covers other elements.

To bring new 2D elements into the deep space, there is the Image2Deep node, which is attached behind all possible 2D nodes such as graphics, footage and renderings. The result is then available in the deepcomp, but of course remains flat itself.
For example, texts can be perfectly integrated into the fireball. Partially concealed, without a mask and the resulting alpha problems.

The final DeepNode is called Deep2Image and brings the deepcomp into the classic 2D space for further processing or output. We have the fireball, but the right reflections are still missing. Let’s take a look at the brand new EXR multilayer workflow.

EXR Multilayer Workflow

For the greatest possible flexibility in compositing CGI renderings, other helpful render layers such as masks, reflections, depth, separate lights and world position are added alongside the actual beauty pass (“the image”), which then allow a number of far-reaching adjustments to be made in compositing.
For reasons of clarity and working speed, these passes, which can be imagined as Photoshop layers, are not rendered separately as individual file sequences, but in a multilayer ERX sequence. This is then split up in compositing – which was also possible in principle in the past, but was very cumbersome.

For Fusion 20, Blackmagic has revised the entire AOV / multilayer system and created a very fast workflow. The first major innovation is the layer dropdown in the viewer, which can be used to view all the layers contained in the EXR or PSD (shortcut: pageUp/Down) – to simplify things, we are using a rendering here that already has the fireball integrated.

At first glance, there does not appear to be an equivalent to Nuke’s shuffle node for extracting the AOVs. However, to get to the individual layers, we can simply use … any node. For the sake of clarity, we use the ChannelBool node, which can generally be used to swap image channels.
(Update with Beta 4: Blackmagic added a Node called Swizzler which acts more like the shuffle node.)

Below on the left viewer we have only extracted the effects of the AOV rimlight on the reflections (LPE rendered from Karma) using Setting > Backgroundlayer and then made them brighter using a ColourCorrector.

In order to return these to the actual image, we must first remove the rimlight present in the beauty pass, otherwise it will be calculated twice. To do this, we can use the newer Multimerge Node, which allows us to process different layers with different mathematical operations. The layer called “Difference” removes the original rimlight, while the “Add Rim” Layer adds the new processed light using Alphagain = 0 (which is the same as the Nuke plus operation).

Caution: The title “Layerlist” only refers to the sequence in the merge node, not to the EXR layers/AOVs themselves.

As the nodes can access the AOVs directly, nothing needs to be extracted for simpler operations such as a Hueshift. The colour corrector can access the layer directly, in this case the emission pass. The rest happens as usual, a merge removes the previous emission pass, a second merge adds the new, blue-coloured pass.

Further application examples: The books on the left are too shiny? Simply subtract the reflection layer (limited to the books by crypto)!

To give only the explosion a glow, the “DirectEmission” pass can be selected directly in the X-Glow Node as an area mask. Last but not least, we add a camera blur using fresh air Depth of Field. Conveniently, the depth map can also be read directly from the stream here.

By the way, if you want to add new layers to the flow, you can do this with the LayerMux node. LayerRegex allows layers to be removed or renamed. The new workflow not only supports EXR, but also Photoshop layers.

What about ML?

Blackmagic has always been well positioned when it comes to machine learning tools. First and foremost the extremely practical Magic Mask, which can significantly speed up rotoscoping tasks. Version 2.0 is already integrated in Resolve 20, while Fusion still has to make do with version 1.0 in the current beta.

In the Magic Mask Node, lines are simply drawn on a reference frame over the objects to be masked and then tracked. The mask itself can be refined or blurred using the “Matte” tab. As the whole operation is quite computationally intensive, the result is best cached (by right-clicking on the node).

To whet your appetite, let’s take a look at the new version, which has been simplified and improved in terms of precision. With our woolly friend, 2 clicks instead of strokes (the sun collector is precisely recognised as a separate object) are enough to create the mask. After tracking, a temporally stable mask is ready for use.

Intellitracker

Another well-functioning machine-learning tool is the Intellitracker, which copes well with even the most jagged movements such as that of this flower in the wind – as can be seen from the movement path aka the green wild line. The Intellitracker is the new default tracker and is automatically selected as soon as the tracking node is called up.

It is usually not necessary to do more than move the tracker to the desired area and press “Track Forward then Reverse”. The tracker automatically selects the most appropriate colour channel, in this case the red channel (see bar chart below the tracker list). In this case, the text was not attached directly to the tracker. A downstream transform node, whose coordinates are linked to the tracker position, offers the flexibility to position the text anywhere in the image. The tracking path can be adjusted directly in the viewport frame by frame like a normal spline or in the Spline(Curve) Editor.

Let’s continue tracking, even without ML tools – with the Surface Tracker. This practical node is able to track intrinsically deforming objects such as clothing, newspapers or flower petals in the wind with a kind of fine grid and transfer the movement of the individual points to any graphics or text so that they follow the complex deformation. In this example, the text was only positioned on a reference frame using classic gridwarp; all other deformations are handled by the surface tracker.

For the sake of completeness, the planar tracker should also be mentioned – neither new nor ML, but tried and tested and easy to use: draw a spline around the area to be tracked, set the operation to e.g. Cornerpin, adjust to the area and define the new insert as the tracker foreground.

Multitext

One node to … surprise them all with beautiful typography. You can think of this innovation a bit like any number of classic text layers, with all the trimmings: transparencies, transitions, extensive typographical settings such as (manual) kerning, modifiers (more on this in a few staggered lines). This was also possible in the past, but a separate node was required for each text element. However, the dynamic frame and circle text familiar from DTP has never been available.

To quickly create procedural animations, you can of course write expressions (more on this later) and link parameters, but you can also add a modifier to the respective parameters by right-clicking and thus quickly create procedural effects such as random movement (wiggle sends its regards).

The modifiers are not just limited to the text node, but can be applied to almost all parameters in almost all nodes (and also linked to each other). For example, a master random modifier can control the opacity of different merge nodes at the same time.

The modifier “Follower”, responsible for “character-by-character” animation, is exciting for moving typographers. This allows you to animate letter by letter in opacity, colour, size, etc. – not only in 2D, but also in 3D, as Fusion has an extensive 3D text node, including bevel and extrude.

Shape Tools

Primarily intended for motion graphics, the Shape Tools are characterised by a vector-like workflow. In contrast to the other 2D systems from Fusion, they are basically resolution-independent and are only cast in pixels using a shape render node.

With the shape tools, various basic shapes and paths can be drawn, combined, duplicated, created as a grid and, of course, animated. Linking individual parameters with the above-mentioned modifiers offers great possibilities. The jitter node ensures random movements. For even more control (or chaos), the shape can be instantiated directly on a particle system.

Expressions

Many motion (and VFX) tasks can be significantly accelerated or automated with simple expressions. To access the expression editor, simply right-click on the relevant parameter field > Expression. For example, as the simplest of all possibilities, values can be changed by the pure passage of time(value “Time”) or linked and nested in complex ways. Parameters can be linked interactively using the small “+” icon.

You can find more examples here:
https://www.steakunderwater.com/VFXPedia/96.0.243.189/index4aa9.html?title=Simple_Expressions

Of masks and multipoly

The aim of the various ways of creating a mask is always the same – to create an alpha channel. Fusion is quite flexible in its use of masks, which are created using rectangles, ellipses, polygon splines or paint nodes. The individual masks can be easily combined, animated and attached to trackers. Individual mask points (if created by path) can be “published” and the parameters of other nodes can be linked to them. The bitmap node converts individual colour channels or the luminance of footage or graphics into masks (more precisely: into an alpha channel). Thanks to the new layer system, AOVs can now also be selected directly as masks here. Thanks to the node system, a mask can be reused or instantiated (copy>shift-v) as often as required.

A few examples:

Above: A simple vignette with ellipse mask node and strongly blurred mask edge aka softedge. In the right viewer the alpha channel of the mask.

Complex shapes are possible by linking individual mask nodes. Alternatively, the newer Multipoly tool can also be used for this, although this is limited to polygon and BSpline masks.

Quick and easy thanks to the new layer system: the blue channel of the AOV glossy transimission as a mask for a colour corrector. Connect the footage loader to the ColorCorrectNode as an input AND as a mask (blue input), set the effect mask layer to the desired AOV under Settings and select a channel under Channel if necessary.

A simple polygon mask in “DoublePoly” mode. The outer outline defines a soft edge gradient, which can be set in addition to the global “soft edge”.

Timesaver: The Multiframe option allows you to change mask points for all keyframes simultaneously – similar to Mocha’s Überkey.

Masks can of course also be drawn beautifully and even support graphic tablet pen pressure …

… and brushes such as this useful fish (available as a preset).

To track a mask, the mask centre can be linked to a tracker modifier. The actual mask path can then still be adjusted or even animated, as only the centre position of the node is actually attached to the tracker.

Particles

Fusion comes with a very powerful and intuitive native particle system. The setup is very simple: The basis for every ParticleFX is the pEmitter Node. As a starting point, it can assume various basic shapes, use 3D shapes (see text example above) or an image input (see crypto example above) and the pRenderNode, which displays our particle system either in 3D space or as 2D pixels.

The particles created in this way can now be subjected to various forces, e.g. the extremely popular turbulence. The effect strength can be set using a 3D mask (region), probability, particle groups or particle age.

Although the particles are not really simulated in comparison to Houdini, they can mimic many effects such as gravity and bouncing of objects

One of the most powerful nodes is the “replicate3D”, which can instantiate any 3D objects on the points and vary their size, rotation and position randomly.

ACES 2.0

Fusion 20 now supports ACES 2.0 & OCIO 2.4.2. Let’s take a look at what this looks like in practice.

The loaded rendering was created in ACEScg, which can be seen in the metadata (hotkey “V”, right-click and display metadata). The display is too dark without display transform.

The small raster icon (1) leads to the Display LUT/Transform menu. Here we select ACES Transform (2) and edit (3) the input & output transform (4) depending on the pipeline, here ACEScg in, sRGB Gamma 2.2 out.

Now we see the image correctly displayed in the viewers (but only displayed, the image itself is not sRGB!) and can continue to perform all operations in ACEScg.
For the final output as e.g. ProRes 4444, however, we have to apply these values via ACES Transform Node.

If you would like to work with AGX / Filmic and co, proceed as follows: Again via the ViewLut menu (1), this time select OCIO Display (2) and edit with the following settings (4). The conversion for the final output is carried out this time via the OCIO Colourspace or Ocio Display (!) node, with the same values as in the ViewLut. The necessary config comes this time from: https://github.com/Joegenco/PixelManager

So much for the application in Fusion. If you want to read more about ACES or argue about it, you can do the former here: https: //chrisbrejon.com/cg-cinematography/chapter-1-5-academy-color-encoding-system-aces

Vector Warp

The new vector tools are based on analysing the movement of the image pixels. Fusion knows what happens to which pixel and can therefore apply complex deformations or retouching quickly and usefully. Areas of application include digital make-up or the insertion of new objects.

The basic prerequisite is motion vectors, which can be supplied externally or generated via an optical flow node (cache without any need!).

To simply place new objects on the background, the new VectorWarp node in “Generate Warp Map” mode is sufficient. The deformed result is placed over the background again using the Merge Node.

For more complex retouching, the VectorWarp node can be set to “Unwarp” and “freezes” the object in time. In this way, objects can be retouched or new ones added using the Paint Node. The result then flows into a 2nd Vector Warp Node, which brings the image back into motion using “Generate Warp Map”.

Curve Warp

The warper previously only available in Fusion Reoslve can now also be controlled using curves. Simply draw a line or an outline, set limits if necessary and bring the object into the desired shape.

Node versioning

A very small and extremely practical function is the somewhat hidden versioning of a node. Up to 6 different settings can be saved to quickly try out different looks. These are not presets, but can be created as such by right-clicking on the node name and Save Settings.

Performance optimisation

Fusion offers various optimisation options to keep performance high despite complex effects and comps:

1. The render area selection (region of interest) limits the calculation to a selectable part of the image
2. Proxy mode reduces the preview resolution (click on the PRX button on the right for further options)
3. High Quality Preview and Motionblur can be switched off for faster previews
4. Framestep skips every frame and leads to faster previews (right click on the play icon for more options)
5. Cache to disc renders and saves the flow up to the selected node (right-click on the desired node)

Reactor

Reactor, which can be downloaded free of charge from the unofficial official Fusion forum, is the counterpart to Nukepedia and lets you install all kinds of macros, scripts and fuses, also commonly known as user-created plug-ins, directly from within Fusion. From exponential glows, edge blur and complete mograph solutions (Krokodove!), everything is included – including the Nuke2Fusion project, which bases shortcuts and settings on Nuke as far as possible.

Installation and instructions: https://www.steakunderwater.com/wesuckless/viewtopic.php?t=2159

Top nodes to try:
X-Glow for wonderful exponential glow
FTL tools for modular lens flares
Krokodove adds many motion graphics nodes to Fusion
OIDN denoiser to be able to use the Intel Open Image Denoiser directly in Fusion.

Last news but not least news: Fusion inside Resolve only: The option you have been waiting for since the integration in Resolve: You can now display the grading applied in the colour page in Fusion and set the start frame count independently of the footage.

Fusion Resolve (Studio) integrated vs. Fusion Studio Standalone

A quick look at the possible versions and versions of Fusion – the standalone is only (still) available as Studio and therefore costs just €355. However, for that money you not only get Fusion, but also Resolve Studio. Or vice versa.
The free Resolve version has Fusion integrated, but has to do without a few really practical Studio OpenFX such as Lens Blur, Termporal Denoise and the new Neural Engine FX such as Magic Mask II. For this reason, the Studio version is highly recommended, especially because it is a perpetual licence. No subscription. For Resolve Fusion. As you can see, it’s worth it.

If you don’t need the other Resolve tools for your current task, it’s better to use the standalone version for performance reasons – it’s simply faster and more flexible as it doesn’t have the Resolve overhead and also offers network rendering (with unlimited render clients). By the way: Before Blackmagic times Fusion alone cost around 2500 $ …

Conclusion

Fusion offers a powerful complete package for compositing / visual effects and also a lot of core power for motion graphics, if you can get involved with the NodeSystem and do without direct integration of Adobe Illustrator files. Working with the programme is fun and quick. The price of the programme is unbeatable, and Resolve Free is even free – just give it a try.

The beta is available immediately and can be downloaded from the Blackmagic website. A Resolve or Fusion dongle or key is required for operation. This is available as a one-off purchase for 355€ – as a perpetual licence.

The post Fusion 20: a deep look at the core first appeared on DIGITAL PRODUCTION and was written by Manuel Kotulla.

The latest release of PFTrack 24.12.19 from The Pixel Farm brings significant updates to its scene-reconstruction capabilities, aiming to enhance efficiency and accuracy for VFX artists. This version focuses on overhauling the photogrammetry toolset, integrating machine-learning technologies to improve performance.

AI-Powered Photogrammetry Enhancements

The Photo Survey node now features a machine-learning-based feature matching engine, designed to be faster and more accurate than previous iterations. This enhancement is expected to streamline the process of reconstructing 3D scenes from still images or video footage, reducing manual intervention and potential errors.

Lidar-Quality Point Clouds

Updates to the Point Cloud node enable the generation of point clouds with Lidar-like quality. This improvement facilitates more precise scene reconstructions, providing a solid foundation for subsequent VFX work. Additionally, new tools have been introduced for cleaning up point clouds by manually removing unwanted points, offering artists greater control over their 3D data.

Workflow and Performance Improvements

The Match Camera functionality has received workflow and performance enhancements, contributing to a more efficient match-moving process. These updates are designed to reduce the time and effort required to align virtual cameras with live-action footage, a critical task in integrating CGI elements seamlessly.

USD Workflow Updates

PFTrack 24.12.19 also includes updates to USD (Universal Scene Description) workflows, with support for point-based geometry. This addition enhances compatibility with other digital content creation tools that utilize the USD format, promoting a more integrated pipeline for complex projects.

Pricing and Availability

PFTrack operates on a rental-only model. Personal subscriptions are available at £25/month (approximately $32/month). Enterprise subscription pricing is available upon inquiry through The Pixel Farm’s website, with reseller prices listed at around $1,125/year/seat. But contact them, because we didn’t find exact prices. Also, there are rumors of a 35 Dollars per month “Personal” License, which re available as an In-App-Purchase. Resellers are (Worldwide) toolfarm.com and in Germany our friends from DVE As

Stability Considerations

While these new features offer promising enhancements, it’s advisable to test PFTrack 24.12.19 in a controlled environment before deploying it in active production. Ensuring compatibility and stability within your specific workflow can prevent potential disruptions during critical project phases. A trial version is of course available.

For a visual overview of the new features, you can watch the official release video below:

The post PFTrack 24.12.19: Scene Reconstruction Gets a Makeover first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

Boris FX has unveiled Mocha Pro 2025, introducing improved planar tracking tools, USD file support, and enhancements to rotoscoping workflows. The latest version of the popular motion tracking and VFX tool focuses on production-friendly updates for compositors, editors, and VFX artists.

Open USD Integration: File Exchange Simplified

Mocha Pro 2025 now supports the USD file format (Universal Scene Description). This widens compatibility for data exchange across modern pipelines, particularly in VFX and animation. Users can import and export USD files to streamline workflows with other USD-supported applications.

Planar Tracking Gets Faster and Smarter

The core feature, planar tracking, remains a focus with speed and stability improvements. While no groundbreaking changes were listed, refinements ensure smoother motion tracking results in a shorter time—good news for artists on tight schedules. Think of it as a “SynthEyes Light” for quick tracks. For more complex shots, the “Full-Fat” SynthEyes remains one of the best trackers out there, but Mocha Pro 2025 offers a solid option for fast and reliable planar tracking within the Boris FX suite. Building on the recent addition of SynthEyes 3D tracking technology, Mocha Pro 2025 now adds 3D moving object solves, allowing artists to solve for moving objects within 3D scenes. By combining planar and PowerMesh tracking data, users can generate accurate meshes and point clouds for complex 3D tracking shots. Solve data integrates seamlessly with popular VFX applications via USD, Alembic, and FBX exports. After Effects users can export solve data in the Mocha Pro UI or generate it directly inside After Effects with the Mocha Pro plugin.

Rotoscoping: Tweaks to Streamline Precision

Mocha Pro’s rotoscoping tools receive practical upgrades, optimizing the precision of spline-based masking. Enhancements aim to simplify shape management while keeping the interface familiar for existing users.

OpenColorIO for Color Management

With OpenColorIO (OCIO) support, artists can match project color spaces between Mocha Pro and other software like Nuke, Resolve, or Blender. This is particularly useful when working on VFX-heavy projects requiring accurate color consistency across the pipeline.

Streamlined Data Exports

A redesigned export dialogue allows artists to export Mocha Pro tracking data more easily using a sleek interface that combines tracking, shape, and 3D data exports in one place. Users can select data export options by popular host application presets, save favorites, and work faster with an “Export Again” feature. Since the current version only offers a dropdown menu, this is an extremely useful update to the “bread and butter” workflow—a small but appreciated quality-of-life improvement for artists.

Compatible and Ready

As always, Mocha Pro integrates seamlessly with leading compositing and editing software, including Adobe After Effects, Premiere Pro, Nuke, and Avid. Licensing options remain flexible, with Mocha Pro available as a standalone application or as a plug-in for host software.

For pricing, head directly to the Boris FX website. (They have a free trial, if you want to see and test it for yourself) As always, test new features thoroughly before deploying them in live projects.

The post Mocha Pro 2025: Planar Tracking, USD, and Roto first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

Working in 2024 has changed significantly for many in the last 5 years. Today, almost 25 per cent (40 per cent of the self-employed – see is.gd/homeoffice_statis and is.gd/hybrid_office) work fully or partially from home. The pandemic in 2020 has therefore doubled the number of people working from home and significantly changed the way we will work in the future: With a headset and webcam in the kitchen. It is therefore essential to be clearly understood and seen in the digital world.

When I bought a powerful Dell XPS 15 in 2023, I noticed one thing: Although the integrated webcam was invisibly built into the monitor, it was also completely useless, either professionally or privately. A surprise for a relatively new and high-quality model and since Dell actually manufactures its own webcams. I bought a Logitech FullHD webcam as a replacement. A significant improvement in comparison, but still not up to scratch. As I look after customers and take part in events and workshops that are also recorded for posterity, the FullHD quality is no longer good enough for me after a while. This led me into the world of 4K webcams and this is where Obsbot Tiny 2 with the gimbal came into play.

How does the Obsbot Tiny 2 work?

The Obsbot Tiny 2 webcam works just like any other webcam on the market: simply plug it in and off you go. First, we connect the station to the webcam itself using a magnet, then we attach the station to our laptop, place it on the table, screw it to a lens using a ¼-inch thread (upside down if you prefer) or attach it to any metal surface. Finally, we connect the webcam to the laptop using the USB-C cable and we’re ready to go.

The moment you take part in an online conference, the Obsbot switches on automatically, which is indicated by a green light and a forward rotation. In idle mode, the light switches off again and the camera turns its lens down towards the station, which we can also do manually – thus switching off the webcam. The great thing about this is that there is no misunderstanding when the webcam is active, so no unwanted video recordings can be made.

Gesture control and gimbal

When the webcam is switched on, we can show it our open palm, which activates automatic tracking; we can switch tracking off again with the same movement. From now on, the Obsbot webcam will automatically follow our face and frame it in the centre, even if we move around the room, as it can rotate through the gimbal. In addition, we can use either gesture or voice control (“Zoom in Closer”) to zoom in and out. The Obsbot app also offers a wide range of setting options for adjusting the zoom, image, HDR on/off and sound.

If you want a few beauty filters, a blurred background or to play with a few colour modes, you need to activate the “Virtual Cam”. This means that you have to install the software and activate this “camera” – so it’s no longer plug and play. However, if you’re working in a fixed location and want to remove the Christmas pounds, you can just slap it on!

Accessories

When purchasing the Obsbot Tiny 2, we receive the webcam with a magnetic station to attach it to the table or monitor. A USB-C to USB-C cable is included to power the device. The valuable camera and accessories can be safely stowed away in the robust carrying case – ideal for travelling. We can download the Obsbot software free of charge from the website, which provides us with additional functions and settings, but is not necessary for operating the webcam. We can also order a remote control separately, which we can use to control the webcam remotely, which can be particularly useful for presentations. Please note that the remote control is supplied without batteries.

Remote control

The remote control is a little more old-fashioned and is probably best suited for standing presentations and for controlling certain aspects. One advantage is that – for example, presentations in a “teaching room” – you can save a few fixed angles and zoom factors – for example: teacher at the lectern at position 1, the blackboard at position 2, and the entire room at position 3. The advantage of a fixed setup is of course that you can label the remote control, and then hopefully even the wobbly chair will still be able to cope with it – for example, if you simply want to switch between teacher, graphics tablet and “painting station” for an animation workshop… Incidentally, all features can also be assigned to system-wide shortcuts within the software, or can be addressed by Autohotkey and the like – keyword Streamdeck! Incidentally, the Tiny 2 can also be addressed via OSC – and can therefore be controlled with the usual event tools. That’s nice with the Tiny 2 – but a future test of the NDI version will be really exciting.

Not so great

The first thing that strikes you when you see the Obsbot is its robot-like design with the gimbal, the second is the price: currently €360 on Amazon is not exactly a bargain. The Obsbot allows gesture control, with which you can zoom in and out with a few simple gestures while talking, which is often more awkward than it sounds. Not only did it happen to me a few times that I had suddenly zoomed in and out of my face during a conversation – which resulted in a moment of confusion – I also couldn’t get away from it with the gestures, not without calling up the app. Gesture control also quickly comes into conflict with popular video conferencing programmes such as Google Meet and Zoom, where holding your hand flat is synonymous with raising your hand. “Do you have a question, Alexander?” is a quick response when you try to switch off the gesture control.

The webcam is equipped with an omnidirectional microphone that enables all-round recording. What sounds good on paper is rather annoying in practice, as background noise is also amplified in most cases. The sound is bass-heavy, but also reverberant and with lots of highs and pops, which is probably due to the denoiser. If you want to hear how it sounds, you can listen to a comparison MP3 here: is.gd/obsbot_audio It is noticeable, but not annoying, that the Obsbot webcam gets noticeably warm after each session. During my test, the camera only failed once, but could be restarted immediately. It is unclear whether this was due to overheating or a programme error. The cable is fine and has a good length, but is a little stiff and doesn’t bend very well. All in all, all the negative points are minor compared to the product itself. However, when you have a 360 euro product, you also have to take care of the details that come with the price.

Conclusion

Having been able to test the Obsbot Tiny 2 extensively over several months, both professionally and privately, my conclusion is: 4K webcams are a good thing and the Obsbot Tiny 2 cuts a very fine figure. An alternative, especially for professionals, would be a 4K webcam without AI tracking, which is of course easy on the budget but lacks some functions. The 4K quality is also noticeable in Full HD calls and recordings and not only makes us look better, but also more professional.

The AI tracking and zoom functions are great for presentations, but also for private conversations where the camera follows us around the room. This can change the way we interact during calls as we are no longer just sitting in front of the device but mobile and interacting with the environment during the call; this also comes at a price.

To summarise, I would recommend upgrading to 4K if budget allows and you want to look better on personal and business calls. Also, if you like the AI and tracking aspects, then Obsbot is definitely worth serious consideration. Alternatives like the Logitech RALLY CAMERA come in at an incredible – is there gold in it? – 1399 euros (Amazon: 976 euros). If you want to buy an Obsbot webcam, I would prefer the Obsbot Tiny 4K to the Tiny 2, as the technical differences are relatively small, but the price difference of 90 euros is relatively large.

The post Obsbot: 4K, CI, Gimbal, .. first appeared on DIGITAL PRODUCTION and was written by Alexander Richter.

The programme was developed largely single-handedly by Dr Russell Andersson over more than 20 years and has long been established not only in the film industry, but also in architecture and forensics.

It has now been incorporated into the BorisFX portfolio. As with Mocha Pro – originally from Imagineer – you can expect this programme to be maintained and developed further. Compared to the remaining competitors 3DEqualiser and PFTrack, the price is quite attractive (Boujou disappeared a few years ago). A good reason for a more comprehensive test!

Interface and operation

At first glance, the GUI looks a little old-fashioned, but don’t let that fool you. The fact that the drop-down menus are usually not sticky and you have to hold the mouse button down until you select the desired function can also be a little irritating. You should forget about the option to switch to languages other than English. It is based on an AI that sometimes makes you go “Huh, what’s that, please?” and where German text often doesn’t fit into the field. The assistant called “Synthia” is also not very helpful at first glance and initially responds stubbornly with “Sorry, I don’t understand.” This is not Siri or Alexa, but rather an assistant for comprehensive automation through scripts in SynthEyes with defined commands. Russ shows how this works here: is.gd/automationsyntheyes.

But that’s it for the criticism, because there is not only a manual with over 800 pages in English, but also various video tutorials that Russ himself has published over the years. A few of the most important ones have already been published on his own YT channel “Boris FX Learn”, even if Russ’s are a little older. This one on masking moving objects using rotoscoping should be very useful for many: is.gd/rotomasking, or this one on mesh building: is.gd/meshbuilding. New additions are is.gd/solving_AE and is.gd/lens_AE, which help with getting started and transferring to After Effects.

In addition to the actual manual, there are nine further PDFs on specific topics such as Planar Tracking or Camera Calibration under “Help”. Comprehensive tool tips and warnings if you make mistakes complete the whole thing. Otherwise, the wealth of functions with 13 tabs and many hidden, additional windows may seem overwhelming at first glance, but right under the first tab “Summary” there is a large, green button called “Auto” at the top left. You can try it out at the beginning, even if experienced professionals turn up their noses at it. Nothing prevents you from refining the results yourself. We also unleashed this function on our more than 20 test clips to see how difficult they were for the programme. There was only one that was not successfully calculated, even if the automatic function sometimes failed to achieve the magic value for precision below 1.0 pixels straight away.

Performance

Precision and speed are of course the most important criteria in everyday production. There are three steps in the process: The identification of suitable image points for tracking, here called “blips”, their tracking over a number of frames as 2D tracking and finally the “solver”, which is the calculation of the spatial relationships for the camera and the scene. Anyone who has ever worked with a point tracker will know what suitable shots should look like: lots of depth of field, little motion blur and good contrasts. For 3D, the parallax of a freely moving camera must also be taken into account.

SynthEyes can also process panning shots from a tripod, but then you only get a circular horizon. We used shots from a drone with a relatively small sensor, hand-held shots with an iPhone 15 and some from a Sony A7IV with a wide-angle lens, also hand-held. As experience has shown that even simple trackers can cope well with shots from an urban environment, we didn’t want to make it so easy for the professional software. We mainly flew over cultivated fields with repetitive structures and natural shapes such as trees, bushes and water, with a few houses thrown in for good measure. Simple trackers usually have difficulties with such motifs.

Not so SynthEyes: With the exception of one, all clips were successfully calculated with “Auto” and could then be brought to values below 1.0 with just a few interventions, in the majority we even came close to 0.5 without much “manual work”. The duration of the clips was between just under one and five minutes, the resolution was UHD. All clips took less than their own runtime on a modest MacBook M1 Pro. The longest, which primarily contained trees, flowing water and sky, was calculated in 3:45, with around 7,500 images. The precision was 0.75 at the first attempt without any intervention. That is clearly professional level!

The 10 CPU cores were fully occupied, on a computer with more “steam” the programme would hardly leave time for a sip of coffee. For comparison: The camera tracker in Fusion took over 13 minutes with the standard setting for a clip of 1:30, but achieved a very good precision of 0.23. After Effects took over 20 minutes for the same clip and achieved 0.47 pixels. These are good values, but with these times you wouldn’t want to do much fine-tuning if a scene is more difficult. In this case, SynthEyes took a good minute and achieved 0.84, but offers plenty of potential for fine-tuning.

Lens data

With early 3D trackers, it was a common recommendation to enter the exact size of the chip and the focal length of the lens. Unfortunately, however, lens manufacturers are not always very precise with the focal length. It is even more difficult today with the chip size. The absolute size can usually be found out, but it is much more difficult with the actual area used. With CMOS sensors, a smaller section is often used, either for higher image frequencies or for internal lens correction. It is usually not possible to find precise information on this.

For our drone, for example, DJI specifies a field of view of 84 degrees diagonally and gives a full-frame equivalent for the focal length of 24mm. However, this applies to photos on the full chip area with an aspect ratio of 4:3. The chip area used for video in UHD is almost impossible to determine, as the small lens is certainly also used for calculations – in any case, the image hardly shows any distortion. With the iPhone, it is generally the case that lens errors are factored out so that the real values are also largely unknown here. It is therefore very helpful that SynthEyes can calculate the necessary distortion correction itself. In the case of the drone, it arrives at just under 70 degrees, which is not unrealistic.

In order not to disturb this calculation, you should never activate internal stabilisation. Both the shifting of the image on the chip and the optical stabilisation, where the entire sensor is moved, would constantly shift the centre of the lens. Mechanical means, such as a gimbal or Easyrig, are of course permitted because the camera itself moves. Blurring, be it a shallow depth of field or too much motion blur, is also not a good starting point for match moving. It is better to add both based on the tracking information in post-production, and SynthEyes can also take care of the stabilisation.

Equalisation of lenses

SynthEyes offers highly complex algorithms for rectification, Standard Radial 4th-order is recommended for spherical lenses, but there are also algorithms for anamorphic lenses. “Read or Cry” should be taken seriously, because if you use this feature incorrectly, the results will tend to be worse. It is an iterative process in which the parameters should be worked through from top to bottom. To do this, you have to switch from “Automatic” to “Refine” in the “Solver” window at the top left after the first tracking and click “Go!” once after each step. In our 90-second clip, one step took less than a second, and this alone almost halved the deviation to 0.48.

Even a rolling shutter is corrected reasonably well, but is generally not a good prerequisite for camera tracking. With the drone or the iPhone 15, this value was low, but with the full-frame camera, this adjustment still brought quite a lot of improvement. In a further step, you could increase the number of trackers and remove the less good ones, but this was not necessary here. We had no difficulty getting below the magic 1 in any of the test clips.

With anamorphic lenses in particular, the question then arises as to whether we should keep the clip in its corrected form after compositing or undo the correction. As a rule, the distortions are not removed in the final result, but are added again after compositing. SynthEyes also has a solution for this, which can be found under “Lens Workflow”. This means that both options are available, and it is particularly easy to use scripts for After Effects, as shown in the second clip for beginners (see above). In this case, SynthEyes even warns you if you have not taken this aspect into account.

Input and output

In addition to precision and speed, a solution for professionals must of course be very versatile in order to fit into any workflow. Right from the start, we were pleased to see how many video formats are accepted. This even includes material in HEVC 10-bit 4:2:2 from hybrid cameras with up to 8K, which is something that the standalone version of Fusion does not recognise. Of course, ProRes is also no problem, but DNxHR only works in a MOV, not in MXF. Image sequences commonly used in animation and VFX, e.g. EXR, are read without any problems, as are DNG, JPEG or PNG series.

Of course, RAW from Arri, Red or Blackmagic is accepted, but currently no MXF clips from the Arri Alexa S-35 or Sony’s Venice and FX9. If necessary, you can switch the preprocessor in front of it, which is almost like a small grading app, but can also read in and memorise LUTs for adjustment. We used this with the iPhone material for Apple Log. With the D-Cinelike from the drone, which is not a real log, we also tested whether a contrast enhancement would produce better results. Here, however, the low-contrast original immediately delivered slightly better precision.

The list of 25 output formats is really impressive. This ranges from long-lost software such as Shake to practically every important compositing and 3D software today. For After Effects there is a Javascript, for Blender, Cinema4D and some other Python scripts that start the respective software and transfer the entire model including camera and video clip – it couldn’t be more convenient. Meshes that you have created and textured in SynthEyes on the basis of trackers can also be transferred. Not only for a moving camera, but also for moving objects. It is even possible to transfer them to photogrammetry software such as Metashape.

Features

The support does not end with conventional video, SynthEyes can also handle 360-degree VR or stereo 3D, the latter in particular is explained in detail in the manual. It can limit the search area for trackers with a simple chromakeyer, but can also use imported alpha masks for this purpose. Even zoom shots can be used, but are more critical in terms of tracking precision. If data is available for this, you can enter keyframes in the lens equalisation, e.g. for zooms.

Integration?

At this point, it will be exciting to see how the company will deal with the new acquisition. It was already announced in the interview (see DP 23:06) that SynthEyes will also be available as a plug-in, although no time horizon has yet been set. With Mocha Pro, BorisFX has shown that it is not simply expanding its portfolio, but that the strengths of the new software are being utilised in other products. This has also started here, as the superior lens correction can already be found in Mocha 2024. Currently, “Camera Solve” in Mocha is still a little slower, even considerably so for very long scenes.

You can define a ground plane, a coordinate origin and scaling based on a known distance between two tracker points. This already works very well for static scenes with a moving camera; moving or deformable objects are in the works. For problematic scenes, the correction options in SynthEyes are much more comprehensive. In Mocha, you are still dealing with a largely automated process in which you can only change the number of trackers and the blip size.

If a “Power Mesh” was defined before camera tracking, this automatically becomes a 3D mesh. You can export this when set to “Single Mesh”, but so far it is still untextured. The export is not as versatile as in SynthEyes, but in addition to some popular target programmes, the increasingly popular USD format is supported and others are in the works.

More can certainly be expected in this area, as Mocha could teach SynthEyes a thing or two in the creation of tracked patches if the products do not completely merge. For example, masks created directly in SynthEyes are less useful elsewhere as they show hard contours. Their creation is also less elegantly solved than working with Mocha 2024, which has been improved again with “Extrapolate Track”. This means that Mocha tracking no longer stops if a few images need to be skipped. SynthEyes can already deliver trackers to Particle Illusion, but with the new “Inherit Velocity” function in version 2024, it could even add the initial movement to the particles.

In our extensive tests, we didn’t find a single scene that couldn’t be tracked with a little fine-tuning, and there weren’t even any crashes. The annual licence now costs 295 US dollars, for shorter projects the programme is available for 49 US dollars per month, and there is also a permanent licence for 595 US dollars. For comparison: PFTrack costs 1,125.00 per year, 3DEqualiser 65.00 for one week. In addition to the tutorials already mentioned at Boris FX Learn, to which more will surely be added, we can warmly recommend Matt Merkovich at Track VFX.

Comment

With SynthEyes, Boris FX has once again acquired a little piece of cream that may eventually merge with Mocha Pro. It comes with the usual exemplary product documentation including bug tracking and workarounds. Experienced SynthEyes artists would certainly be happy if a “Classic” version were to be retained in the event of any further development of the GUI, as is the case with Mocha.

The post SynthEyes at BorisFX first appeared on DIGITAL PRODUCTION and was written by Uli Plank.

CinCraft Scenario is based on Ncam’s unique tracking technology and Zeiss’ expertise in lens data – Zeiss presents a new and flexible camera tracking system as part of its CinCraft ecosystem.

Zeiss CinCraft Scenario provides camera tracking data for use in real-time rendering engines for virtual production or live compositing. Users can also record the tracking data for post-production: “It is well known that camera tracking is extremely useful for pre-visualisation and virtual production. CinCraft Scenario is particularly useful for VFXs as the system records the tracking data on set for use in post-production.

The camera tracking system can be used in almost all conditions and setups – indoors and outdoors, on film sets, with bluescreens or greenscreens and in LED volumes. This is made possible by the hybrid tracking technology, which can detect not only industry-standard reflective markers, but also environmental features such as object edges. CinCraft Scenario also works with digital markers on LED walls that are invisible to the main camera.

CinCraft lens data

As part of the CinCraft ecosystem, Scenario accesses the Zeiss lens database – initially with Zeis and Arri/ZEiss cine lenses, other standards and systems may soon follow. Their integration makes the system “plug-and-play” capable and avoids the tedious process of lens mapping and calibration.

Modular components

CinCraft Scenario works with commercially available equipment and is compatible with the most important camera and lens encoder systems to provide all the data required for correct tracking. The system consists of various hardware and software components that can be customised to meet the individual requirements of each user:

– CinCraft CamBar records raw data from the environment and forwards it.
– CinCraft Origin, which runs the main software, aggregates the data to calculate position and movement
– CinCraft Link connects CamBar and Origin and enables access to the user interface via mobile devices.

The modular software components make it possible to calculate position and movement, record data, export the recorded data to industry-standard file formats and integrate it into the Unreal Engine in real time.

Availability

Users can configure the CinCraft Scenario Camera Tracking System according to their requirements or select one of four pre-configured bundles. The system will be available at selected retailers and via the Zeiss CinCraft website in early Q4 2023.

For more information, visit zeiss.ly/cincraft-scenario.

The post Cincraft Scenario is coming! first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

Lens manufacturer Zeiss has acquired Ncam – the British specialist for virtual production / camera tracking is set to bring expertise for VFX workflows in-house. Incidentally, Ncam was honoured with the “Emmy Award for Outstanding Achievement in Engineering Development” for its technology in 2016 – and Ncam’s hybrid camera tracking technology makes it possible to spatially locate the camera on set – both indoors and outdoors. This data is essential for compositing (live and offline) and, of course, for virtual production. If you want to know more – we’ve featured Ncam several times in this magazine. Here are two of the articles for free download:

DP2105_Ncam

Ncam in use at “Face the Darknes”

Zeiss, who some time ago were looking in the direction of VFX/compositing with the CineCraft system(here is the article from DP 22:05), are thus expanding their expertise – and perhaps everything from the lens to the projector will soon come from a single source? We’ll keep you up to date!

The post Zeiss takes over NCam first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

In nuce: The user Xavier (Jack), developer, animator and programmer in one, has presented the current development status of his real-time hand tracking app via Twitter. On 23 January, Xavier posted a moving image on Twitter that demonstrates what his app currently looks like in use. See below:

Good news!

I got a small grant to experiment with the creative use of real-time hand tracking in the browser.

First step, get that hand rig working with mediapipe (@tensorflow) and @threejs pic.twitter.com/RwXocbxZqJ

— Xavier (Jack) (@KMkota0) January 23, 2023

In toto: The highlight: His app should work directly in the web browser. According to Xavier, his app should still achieve good results even if the fingers being recorded are obscured by objects in the foreground of the image, for example. During development, Xavier utilised both the open source library TensorFlow , which focuses on machine learning for research and production, and Three.js, a cross-browser JavaScript library and application programming interface that can be used to create animated 3D graphics in the web browser.

Click further: To the Twitter account of the user Xavier (Jack).

Source: 80.lv ( message from Theodore McKenzie)

The post The future of real-time hand tracking? first appeared on DIGITAL PRODUCTION and was written by Patrick Poti.

In nuce: Andersson Technologies released SynthEyes 22.10 on 20 October this year – the latest version of the software for 3D tracking and match moving. The newly added flagship functions include a curve-based colour correction tool and improvements in terms of roto masking. There are also numerous quality of life improvements and bug fixes.

RotoMasking in SynthEyes 2210

In toto: According to cgchannel.com, the new curve-based colour correction tool also has animated control points. Roto masking has been improved to the extent that it is now possible to move and rotate rotosplines as a unit or to move several rotospline control points at the same time. In the video above, Russ Andersson shows how roto masking works with SynthEyes 22.10. Furthermore, it is now possible to display a checkerboard pattern. This shows which pixels have been assigned to the current active tracker host by the auto-tracker. The option of superimposing the checkerboard pattern on the view is intended to allow the current setup to be checked. Quality of Life improvements include, for example, a new eyedropper tool that can be used to select the colour for keying.

Key data: SynthEyes 2210 is available for Windows, Linux and Max operating systems. The software ispriced between 299 and 699 US dollars, depending on whether buyers choose the Intro or Pro version and for which licence type. A trial version is also available. Registration is required to take advantage of this. Andersson Technologies was founded by software slash hardware developer Russ Andersson . Russ specialises in 3D animation, computer graphics, robotics and user interfaces for virtual reality and has a doctorate in computer science from the University of Pennsylvania. Russ has worked at Bell Laboratories, the former research department of the American Telephone and Telegraph Company (AT&T).

Click further: The full range of functions of SynthEyes 22.10 can be viewed at ssontech.com . Digital Production reported on the release of Syntheyes 21.02 on 10 February 2021.

Sources: cgchannel.com ( message from Jim Thacker), ssontech.com ( list of functions for SynthEyes 22.10)

The post SynthEyes is released in version 22.10! first appeared on DIGITAL PRODUCTION and was written by Patrick Poti.