For those who do not know the tool, RealityScan is Epic’s photogrammetry application for reconstructing 3D assets, scenes, and datasets from photographs and scans. It is used in areas such as VFX, game development, visualization, heritage capture, and geospatial workflows. RealityScan 2.1, the previous major release, added broader automation options, support for SLAM data, and other workflow extensions, so 2.1.1 reads very much like the expected follow-up pass: less about headline features, more about making interchange and export behave properly when real projects get involved.

RealityScan 2.1.1 is now available, and while this is not the kind of release that arrives with fireworks and a new category of buzzword, it does address several parts of the workflow that tend to matter once scans actually have to leave the application and survive contact with the rest of production.

The most relevant improvements are in the COLMAP pipeline. RealityScan 2.1.1 now exports either to a flat folder structure or to the standard COLMAP directory layout, supports both binary and ASCII export, improves mask export behavior, and filters low-accuracy points automatically during export. On import, Epic has added support for the FULL_OPENCV camera model, which should make it easier to bring in a wider range of calibration data without extra cleanup elsewhere in the chain.

Mask handling is one of the more useful practical changes here. The exporter now supports AI-generated masks, and users can choose whether to keep the .mask.ext suffix when exporting into the standard COLMAP directory structure. Epic also lists fixes for cases where selected mask layers failed to export, along with corrections for quality degradation when exporting COLMAP data based on original distorted images. That is the sort of update that sounds small in a changelog and rather less small when it breaks a handoff.

Another addition is the new “RealityScan XMPs with Image List” export in the Registration exports. This makes it possible to export XMPs corresponding either to the original distorted images or to undistorted images, and to choose between the “local:1 Euclidean” and “Grid plane” coordinate systems. At the same time, Epic has removed the separate Image list export from the Alignment tab, because the same function already existed in Registration export. Exported image lists now use the .imagelist extension, which can be reused when importing inputs or input selections back into RealityScan.

There are also smaller but welcome naming and consistency changes. Texture layer names now respect the specified name, including when selecting layers for texturing. Epic also removed some extra information from those names, including 16-bit and HDR prefixes, and adjusted numbering. On the command line side, importColmap becomes loadColmap, importBundler becomes loadBundler, and importFlightLog becomes importTrajectory. Not glamorous, but coherent. That alone can save someone a mildly irritating half-hour that could be spent watching cat videos.

The fixed-issues section is long, and in this case that is probably good news. Epic lists corrections for an incorrect “model is already colorized” warning on certain LiDAR models, ignored intrinsic parameters during flight log imports, EXR import issues, misplaced imported masks, raster artifacts in depth and mask exports, black mask exports in some cases, crashes related to LOD export settings and long computations, scene transformations being incorrectly retained after COLMAP and Bundler imports, and support for importing PLY files generated by the CGAL library. There are also multiple fixes for export behavior in cameras, orthographic projections, LAS classifications, contour export, and Nira upload handling.

Epic Games documentation for RealityScan 2.1.1

RealityScan 2.1.1 adds better COLMAP support and new XMP export options

RealityScan 2.1.1 is a maintenance release, but a practical one. Epic updates COLMAP exchange, adds new XMP with image list export options, cleans up mask export, and fixes a long list of editor, alignment, and CLI issues.

06.04.2026

Epic Games updates RealityScan: 2.1 out

pic’s RealityScan 2.1 brings LiDAR & SLAM import, server automation tools and improved UV and rendering support.

27.11.2025

RealityScan Mobile 1.8 with Tools for On-the-Go Scanning

Version 1.8 of RealityScan Mobile introduces three capture modes (including AR Guidance), mesh clean-up tools, focus peaking, capture timer and watertight mesh output: ideal for everybody capturing on-the-go.

21.11.2025

RealityScan 2.0: Epic Gives RealityCapture a New Name and New Tricks

Epic overhauls RealityCapture into RealityScan 2.0, complete with AI masking, LiDAR fusion, and a mobile-first workflow focus.

17.06.2025

RealityScan 1.5.3 Update Eases 3D Scans with Better Processing for iOS

Epic’s RealityScan update brings flexible processing, smoother workflows, and better support for iPad Pro and image uploads.

07.11.2024

/

The post RealityScan 2.1.1 adds better COLMAP support and new XMP export options first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool: Atlux λ plugs into Unreal Engine and ships a virtual studio, lights, presets, and automation for fast renders. It is sold on Fab and can also be installed with the Epic Games Launcher.

Fab listing, now with Mac support

Atlux λ is now available on Fab as a visualisation plugin for Unreal Engine. The newest version adds macOS support. The plugin is solo-developed by Jorge Valle Hurtado. The plugin requires account registration. The listing also shows no rating yet, Allows usage with AI set to Yes, and Generated with AI set to No.

The pitch, minus the pitchforks

The core idea: a complete in-editor photo studio-style workflow aimed at speeding up lighting, rendering, and cinematic work inside Unreal Engine. In practice, that means a Virtual Photo Studio, realistic light assets, studio and light presets, camera motion presets, an automated turntable, batch rendering, post-process materials and LUTs, procedural gobos, VFX reference assets, and a simplified render interface.

A key workflow point is automation around sequences and rendering. Rendering takes one click, with everything handled automatically, and no manual Level Sequences or Movie Render Queue setup required. That is a big promise, so treat it as a workflow claim until you validate it in your own scenes and build environment.

One click rendering as a pipeline building block

If you have ever tried to standardise look dev outputs across a team, you know the dull truth: the hardest part is rarely the final pixels. It is the repeatability. Atlux λ leans into repeatability by bundling its own studio environments, lighting assets, presets, and a streamlined render interface, then pushing users toward a single-button render flow. The idea is to keep the shot setup and output consistent, especially when multiple artists work on the same asset library.

That is also where the curated content matters. When a tool ships with production-ready assets and presets, it reduces the number of decisions per shot and lowers the odds that one artist accidentally reinvents the sun at 3 a.m. It also gives supervisors a clearer target for what “correct” looks like in a shared lighting and camera language.

Gaussian Splatting prep, with COLMAP export

The feature list includes COLMAP export for Gaussian Splatting workflows. For teams experimenting with Splatting , the practical value depends on how smoothly the exported data lines up with your existing reconstruction steps and tooling, including COLMAP. The sources do not provide detailed specifications for what is exported or how it maps to downstream tools, so you will need to validate that part yourself.

Atlux λ Live and packaging as a desktop app

The feature list also includes Atlux λ Live, described as a “runtime version” that allows users to package a scene as a fully interactive desktop application, for purposes such as interactive client presentations, sharing scenes without requiring the Unreal Editor, and delivering a controlled runtime experience using the toolset.

If you have ever handed off an interactive review build to a client, you already know why this matters. Packaging is rarely hard in theory. It is hard because it is always the week you cannot afford surprises. Treat the packaging automation as a convenience feature that still deserves a dry run on your target machines and OS versions.

Pricing and licensing, in plain numbers

Pricing is now official: A Personal perpetual license is listed at US$249.99 excluding tax. A Professional perpetual license is listed at US$349.99 excluding tax. Both are perpetual licenses with one active user on any computer, offline access, and full support, updates, and maintenance.

The buy page has a standalone installer for Windows only for the Personal and Professional options, and it also describes purchasing via Fab with the same Personal and Professional pricing, plus availability on Mac and inclusion in the Epic Games Launcher.

Educational pricing is also specified on the same page. An Educational Rental is listed at US$4.95 per month excluding tax, described as cancel any time, one active user on any computer, and support, updates, and maintenance while active, with Windows only and a requirement for school or academy validation.

Support is for Unreal Engine versions 5.4, 5.5, 5.6 and 5.7.

// Fab listing page for Atlux λ
https://www.fab.com/listings/3c321e8e-f0b0-4235-ba38-02d59299869b
// Official product homepage and feature overview
https://atlux.one/

The post VINZI Atlux λ arrives on Fab first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

The Camera Array Tool for Blender, developed by ToppiNappi (Olli Huttunen), generates camera rigs directly from 3D geometry. The add-on places virtual cameras at the centre of each polygon on a selected mesh and orients them according to the surface normal.

Users can decide whether the cameras face inward, outward, or in both directions. The tool allows the creation of custom rigs as well as prebuilt configurations, including grids, spheres, and cylinders. Once generated, the camera positions remain linked to the surface, meaning scaling or editing the underlying mesh automatically updates the rig.

Built for image-based reconstruction

Each camera in an array can render still images or animation sequences, stored in user-defined folders. The output data includes COLMAP text files—cameras.txt, images.txt, and points3D.txt—enabling direct use in 3D Gaussian Splatting (3DGS) or neural rendering workflows without the usual Structure-from-Motion reconstruction step. Point cloud exports can include colour data, accelerating later training phases in external applications. A density slider helps control file size when generating point clouds.

4DGS and animated datasets

Version 3.2.0 introduces 4DGS (Four-Dimensional Gaussian Splatting) data export for animated scenes. The add-on can render each frame from all cameras while generating corresponding COLMAP data per frame. This update fixes a previous COLMAP issue: files are now written immediately after each frame, preventing data loss if rendering is interrupted. A new “Animated Camera” feature enables rendering the same array setup through a single moving camera—an alternative method for producing dataset sequences.

Compatibility and workflow integration

Image series from the Camera Array Tool can be trained into Gaussian Splatting models using external software such as Postshot, NeRF Studio, Brush, or ToppiNappi’s free Postshot Batch Trainer. According to the developer, the add-on is particularly useful for synthetic dataset creation, virtual production, or academic experiments with neural rendering and multi-view reconstruction.

Simplicity over spectacle

Despite its experimental applications, the add-on’s concept remains simple: turn a mesh into a camera grid and render from every face. This direct approach makes it accessible for both research and practical visualisation tasks. The tool costs €36 and is available through Gumroad. The “Virtual Camera Bundle” combines the Camera Array Tool with the 360 Extractor add-on at a discounted rate.

As with all Blender extensions, users are advised to test the plug-in within their own pipeline before committing to production use.

The post Camera Arrays for the Gaussian Age first appeared on DIGITAL PRODUCTION and was written by Bela Beier.