[CVPR 2026] Official implementation of "ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Models"

CoWTracker: Tracking by Warping instead of Correlation

Visual scheme for the Lilliput project

π RuView: WiFi DensePose turns commodity WiFi signals into real-time human pose estimation, vital sign monitoring, and presence detection — all without a single pixel of video.

Official implementation of paper [DeepTag: A General Framework for Fiducial Marker Design and Detection]

This is a lightweight GAN developed for real-time deblurring. The model has a super tiny size and a rapid inference time. The motivation is to boost marker detection in robotic applications, howeve…

Code for the project "MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos"

[RSS'25] This repository is the implementation of "NaVILA: Legged Robot Vision-Language-Action Model for Navigation"

A flexible, high-performance 3D simulator for Embodied AI research.

PyTorch Implementation of MVSNet

A GPU-accelerated TSDF and ESDF library for robots equipped with RGB-D cameras.

MonSter++: A Unified Geometric Foundation Model for Stereo and Multi-View Depth Estimation via the Unleashing of Monodepth Priors

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

RTAB-Map's ROS package.

国内首个占据栅格网络全栈课程《从BEV到Occupancy Network，算法原理与工程实践》，包含端侧部署。Surrounding Semantic Occupancy Perception Course for Autonomous Driving (docs, ppt and source code) 在线课程主页：http://111.229.117.200:8100/ (作者独立搭建)

Depth Pro: Sharp Monocular Metric Depth in Less Than a Second.

Code for robust monocular depth estimation described in "Ranftl et. al., Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer, TPAMI 2022"

[CVPR 2023] An academic alternative to Tesla's occupancy network for autonomous driving.

This repository contains the code for the paper "Occupancy Networks - Learning 3D Reconstruction in Function Space"

GeoCalib: Learning Single-image Calibration with Geometric Optimization (ECCV 2024)

Visual SLAM/odometry package based on NVIDIA-accelerated cuVSLAM

[CVPR 2025] JamMa is a lightweight image matcher that enables fast internal and mutual interaction of images with joint Mamba.

[CVPR 2025 Best Paper Nomination] FoundationStereo: Zero-Shot Stereo Matching

在 ROS 上run的 llama3.1-8b 聊天机器人，无需联网，直接 ollama 本地部署调用！

Joint deep network for feature line detection and description

This code contains an algorithm to compute stereo visual SLAM by using both point and line segment features.

Depth Anything 3

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

Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding