Tiny, no-nonsense, self-contained, Tensorflow and ONNX inference

Scenic: A Jax Library for Computer Vision Research and Beyond

Visual Inertial Odometry using SuperPoint and GTSAM

Release repo for our SLAM Handbook

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

[CVPR 2025 Highlight] Real-time dense scene reconstruction with SLAM3R

The repo for "Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image" and "Metric3Dv2: A Versatile Monocular Geometric Foundation Model..."

Universal Monocular Metric Depth Estimation

kapture is a file format as well as a set of tools for manipulating datasets, and in particular Visual Localization and Structure from Motion data.

Objectron is a dataset of short, object-centric video clips. In addition, the videos also contain AR session metadata including camera poses, sparse point-clouds and planes. In each video, the came…

ONNX-compatible Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data

Efficient neural feature detector and descriptor

🦀 Low-level 3D Computer Vision library in Rust

LightGlue: Local Feature Matching at Light Speed (ICCV 2023)

Visual localization made easy with hloc

Web-based 3D visualization + Python

Variable-resolution mesh extraction from SDFs/occupancy using Octree marching cubes

[3DV25] Official code for "Towards Foundation Models for 3D Vision: How Close Are We?"

HOTA (and other) evaluation metrics for Multi-Object Tracking (MOT).

[NeurIPS 2024] Depth Anything V2. A More Capable Foundation Model for Monocular Depth Estimation

Metric depth estimation from a single image

SuperGlue: Learning Feature Matching with Graph Neural Networks (CVPR 2020, Oral)

CUDA accelerated rasterization of gaussian splatting

A geometry-shader-based, global CUDA sorted high-performance 3D Gaussian Splatting rasterizer. Can achieve a 5-10x speedup in rendering compared to the vanialla diff-gaussian-rasterization.

Ocean is the in-house framework for Computer Vision (CV) and Augmented Reality (AR) applications at Meta. It is platform independent and is mainly implemented in C/C++.

PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

Embree ray tracing kernels repository.