AliceVision is a photogrammetric computer vision framework that provides 3D reconstruction and camera tracking algorithms for converting unordered photographs or video into 3D geometry.
The framework solves the problem of inferring 3D scene geometry from 2D photographs by implementing state-of-the-art computer vision algorithms. It reverses the projection process that occurs when a 3D scene is captured onto a 2D plane, recovering the lost depth information through photogrammetric analysis. The project combines academic research with industrial robustness requirements to deliver production-ready algorithms.
Developers should choose this tool when building applications that require 3D reconstruction from image collections. It suits projects involving structure-from-motion, multiview stereo, camera tracking, panorama stitching, and HDRI image processing. The framework is designed to be used through Meshroom, a separate project that provides both a graphical user interface and command-line tools for launching the AliceVision pipeline. Meshroom's Python foundation allows developers to create custom scripts and automate pipeline steps according to project needs.
The project maintains active continuous integration with automated build testing. Development follows a collaborative model that welcomes contributions meeting quality standards, with discussions and feature requests handled through a public mailing list alongside informal sharing of results and related technologies. The codebase is organized as a recursive Git repository with comprehensive documentation of research papers and tools integrated into the framework.
