Web-based 3D visualization + Python

Public code release associated with SceneScript.

Grounded SAM 2: Ground and Track Anything in Videos with Grounding DINO, Florence-2 and SAM 2

3D Reconstruction for all

Burn is a new comprehensive dynamic Deep Learning Framework built using Rust with extreme flexibility, compute efficiency and portability as its primary goals.

Accurate geometric camera calibration with generic camera models

Optimal Transport Aggregation for Visual Place Recognition

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

Slint is a declarative GUI toolkit to build native user interfaces for Rust, C++, or JavaScript apps.

[ICRA'24] Accurate Prior-centric Monocular Positioning with Offline LiDAR Fusion

Collection of best practices for working with Docker/Docker-Compose and the Robot Operating System (ROS/ROS 2) in simulation as well as with hardware and real-time requirements

🚀 A curated list of awesome articles, videos, and other resources to learn and practice software architecture, patterns, and principles.

Super Repository for Coding Interview Preperation

Implementation of XFeat (CVPR 2024). Do you need robust and fast local feature extraction? You are in the right place!

Template-based implementation of RANSAC and its variants in C++

Stereo Semi Global Matching by cuda

LLM inference in C/C++

C++ Insights - See your source code with the eyes of a compiler

The Kalibr visual-inertial calibration toolbox

COLMAP - Structure-from-Motion and Multi-View Stereo

An open-source C++ library developed and used at Facebook.

Code release for CVPR'24 submission 'OmniGlue'

A monitor of resources

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.

Production-grade 3D gaussian splatting with CPU/GPU support for Windows, Mac and Linux 🚀

[3DV 2025] Code for "FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent" by Cameron Smith*, David Charatan*, Ayush Tewari, and Vincent Sitzmann

[ECCV 2024 - Oral] ACE0 is a learning-based structure-from-motion approach that estimates camera parameters of sets of images by learning a multi-view consistent, implicit scene representation.

[CVPR 2024 Oral] Rethinking Inductive Biases for Surface Normal Estimation

Minimal solvers for calibrated camera pose estimation

open Multiple View Geometry library. Basis for 3D computer vision and Structure from Motion.