Wan: Open and Advanced Large-Scale Video Generative Models

Fast mesh to 3D gaussian splat conversion

MUSt3R: Multi-view Network for Stereo 3D Reconstruction

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

DUSt3R: Geometric 3D Vision Made Easy

[CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

MambaGlue: Fast and Robust Local Feature Matching With Mamba @ ICRA'25

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Code for "MatchAnything: Universal Cross-Modality Image Matching with Large-Scale Pre-Training", Arxiv 2025.

Gaga: Group Any Gaussians via 3D-aware Memory Bank

3DGS-to-PC: 3D Gaussian Splatting to Dense Point Clouds [3D-VAST: ICCVW 2025]

PyTorch3D is FAIR's library of reusable components for deep learning with 3D data

[ICRA'23] BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation

Code for 3D object detection for autonomous driving

(ICLR) Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving

Doppelgangers++: Improved Visual Disambiguation with Geometric 3D Features

Official implementation of "DepthLab: From Partial to Complete"

Image matching with MOP+MiHo+NCC

A collection of papers on neural field-based inverse rendering.

[CVPR 2025 - Spotlight] Official PyTorch implementation of MAtCha Gaussians: Atlas of Charts for High-Quality Geometry and Photorealism From Sparse Views

[NeurIPS 2024 Spotlight] Implementation of the paper "3D Gaussian Splatting as Markov Chain Monte Carlo"

An open source re-implementation of Caesar III

Official repository for the ICCV2023 paper "Kick Back & Relax: Learning to Reconstruct the World by Watching SlowTV"

a programming library with geometric algorithms

Official repository for Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairs

converts image matches to COLMAP format

[ECCV2024] API code for T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy

From Correspondences to Pose: Non-minimal Certifiably Optimal Relative Pose without Disambiguation (CVPR 2024 - Highlight)