[ICLR 2026] π^3: Permutation-Equivariant Visual Geometry Learning

TUI for the Slurm Workload Manager

[CVPR 26] Release repo of our work "Co-Me: Confidence-Guided Token Merging for Visual Geometric Transformers"

Any4D: Unified Feed-Forward Metric 4D Reconstruction

[ICCV 2025 Oral] MVTracker: Multi-view 3D Point Tracking

MapAnything: Universal Feed-Forward Metric 3D Reconstruction

[IROS'25] Source code for "RayFronts: Open-Set Semantic Ray Frontiers for Online Scene Understanding and Exploration"

Open source Structure-from-Motion pipeline

Generalizable Perception Stack for all things 3D, 4D & Scene Understanding

UFM: A Unified Dense Image Correspondence Estimator for both Optical Flow & Wide Baseline Matching Tasks. Matches any pair of images. (NeurIPS 2025)

A PyTorch native platform for training generative AI models

Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)

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

[DEPRECATED] GLOMAP - Global Structured-from-Motion Revisited

Source Code for "Map It Anywhere (MIA): Empowering Bird’s Eye View Mapping using Large-scale Public Data"

VGGSfM: Visual Geometry Grounded Deep Structure From Motion

[CVPR 2024] GLACE: Global Local Accelerated Coordinate Encoding

A comprehensive list of Implicit Representations and NeRF papers relating to Robotics/RL domain, including papers, codes, and related websites

Official repository for "AM-RADIO: Reduce All Domains Into One"

Code for "Cameras as Rays"

Curated list of papers and resources focused on 3D Gaussian Splatting, intended to keep pace with the anticipated surge of research in the coming months.

DUSt3R: Geometric 3D Vision Made Easy

SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM (CVPR 2024)

[ICCV 2023 Oral] ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes

A comprehensive list of Implicit Representations, NeRF and 3D Gaussian Splatting papers relating to SLAM/Robotics domain, including papers, videos, codes, and related websites

A framework for 4D reconstruction from monocular videos.

AnyLoc: Universal Visual Place Recognition (RA-L 2023)

A python library for handling poses, transforms and frames for robotics applications

PyTorch code and models for the DINOv2 self-supervised learning method.

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.