Perception & geometry

Open source

Model-free 6D pose tracking for previously unseen objects.

Open source

DexPoint: Generalizable Point Cloud Reinforcement Learning for Sim-to-Real Dexterous Manipulation, CoRL 2022

Open source

Computer-vision library for image processing, calibration, and perception.

Open source

YOLO toolkit for detection, segmentation, pose estimation, and tracking.

Open source

Source code for the Webots robot simulator.

Open source

3D object detection in point clouds using deep Hough voting.

Open source

A toolbox for learning visual pushing and grasping with RL.

Open source

API for a multi-agent household simulation environment.

Open source

ROS 2 node for AprilTag detection

Open source

A complete end-to-end demonstration in which we collect training data in Unity and use that data to train a deep neural network to predict the pose of a cube. This model is then deployed in a…

Open source

Feed-forward metric 3D reconstruction implementation.

Open source

AI2-THOR source code for interactive visual-AI environments.

Open source

Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learning

Open source

Detects 6-DoF grasp poses in point clouds.

Open source

🐍 Geometric Computer Vision Library for Spatial AI

Open source

An open harness to simulate and deploy perception-based PX4/ArduPilot drone swarms with ROS2, YOLO, LiDAR, NVIDIA Jetson

Open source

Toolkit for quantitatively scoring LeRobot data episodes.

Open source

Multiple-view geometry and structure-from-motion library.

Open source

An open-source, full-stack robotic neck for active perception

Open source

Public implementation of "Beyond Cartesian Representations for Local Descriptors", ICCV 2019

Open source

Generic calibration for robots

Open source

Procedural interactive 3D houses for embodied-AI experiments.

Open source

Toolkit for neural radiance fields and 3D scene reconstruction.

Open source

Frustum PointNets for 3D object detection from RGB-D data.