A general-purpose Structure-from-Motion and Multi-View Stereo pipeline with GUI and command-line interfaces for 3D reconstruction from images.
alicevision/Meshroom
Open-source node-based visual programming tool for photogrammetry, turning photos into 3D models, camera tracking and HDR images.
Open-source alternative to:RealityCapture
Agisoft Metashape
- Stars
- 13,065
- Forks
- 1,242
- Open issues
- 543
- Last commit
- 19 hours ago
What is it?
Meshroom is a node-based visual framework for building and running data processing pipelines. Each node performs one operation, and its output attributes can feed into later steps. When a parameter changes, only the affected downstream nodes are recomputed, while cached intermediate results are reused.
The bundled AliceVision plugin provides pipelines for 3D reconstruction from multi-view photos, camera tracking, HDR fusion from bracketed exposures, panorama stitching including fisheye and motorized heads, and photometric stereo. Additional plugins from MeshroomHub add segmentation, monocular depth estimation, dense feature matching, Gaussian splatting and geolocation.
Execution can run locally or be distributed to render farms. The interface includes a graph editor, per-node logs and statistics, and 2D and 3D viewers for inspecting outputs.
Key features
- 01Node graph where changing one attribute recomputes only downstream nodes.
- 02Photogrammetry pipeline turning multi-view photos into 3D models and camera poses.
- 03HDR fusion and panorama stitching pipelines from AliceVision.
- 04Local or render-farm distributed execution.
- 05Built-in 2D and 3D viewers and per-node logs.
Use cases
- Creating 3D models of objects or buildings from a set of photographs.
- Estimating camera motion from video or image sequences.
- Merging bracketed exposures into HDR images or stitching panoramas.
- Running large batches of reconstruction jobs across a render farm.
Quick start
- 1
Download pre-compiled binaries for the latest release from the releases page
- 2
Or build from source following INSTALL.md
Things to know
- Building from source requires setting up several prerequisites described in INSTALL.md.
- Some advanced plugins are optional and separately maintained.
In collections
Topics
FAQ
- What is Meshroom?
- Open-source node-based visual programming tool for photogrammetry, turning photos into 3D models, camera tracking and HDR images. Meshroom is a node-based visual framework for building and running data processing pipelines. Each node performs one operation, and its output attributes can feed into later steps. When a parameter changes, only the affected downstream nodes are recomputed, while cached intermediate results are reused.
- Is Meshroom free to use? What license does it have?
- Meshroom is open source under the Other license. Check the license terms before using it in commercial products.
- Is Meshroom still maintained?
- Actively maintained. The last commit to Meshroom was on Nov 6, 2026.
- What can I use Meshroom for?
- Creating 3D models of objects or buildings from a set of photographs. Estimating camera motion from video or image sequences. Merging bracketed exposures into HDR images or stitching panoramas. Running large batches of reconstruction jobs across a render farm.
- How do I install Meshroom?
- Download pre-compiled binaries for the latest release from the releases page Or build from source following INSTALL.md
- Where can I get Meshroom?
- The source code and installation instructions are on GitHub: https://github.com/alicevision/Meshroom
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Cross-vendor 3D Gaussian Splatting trainer that turns photos or video into splats and textured meshes, running on Vulkan or CUDA without Python.
A Python package for fast and robust image stitching, building panoramas from overlapping photos on top of OpenCV's stitching module.
Photograph a real object with your phone and get a generated 3D model imported straight into Blender.
An iPhone LiDAR capture app plus Three.js viewer and agent skill for rebuilding rooms as editable 3D scenes.
Official implementation of FlowHMR, a generative model that reconstructs physically plausible 3D human motion from monocular video.