Highlights
- Pro
Lists (2)
Stars
SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning
OKVIS2-X: Open Keyframe-based Visual-Inertial SLAM Configurable with Dense Depth or LiDAR, and GNSS
[CoRL'25] Articulated Object Estimation in the Wild: ArtiPoint + Arti4D Dataset
A high-performance Python-based I/O system for large (and small) deep learning problems, with strong support for PyTorch.
The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that sho…
NVIDIA Isaac GR00T N1.6 - A Foundation Model for Generalist Robots.
VITRA: Scalable Vision-Language-Action Model Pretraining for Robotic Manipulation with Real-Life Human Activity Videos
[ICCV 2025] The official implementation for EgoM2P: Egocentric Multimodal Multitask Pretraining.
A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.
A simple state update rule to enhance length generalization for CUT3R
A cross-environment tool to publish objects from simulation for Augmented Reality and Human Robot Interaction.
MapAnything: Universal Feed-Forward Metric 3D Reconstruction
VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning
[ARXIV’25] Learning Video Generation for Robotic Manipulation with Collaborative Trajectory Control
EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video
[3DV 2026] ViSTA-SLAM: Visual SLAM with Symmetric Two-view Association
Visualize PyTorch tensors with a single line of code.
Reference PyTorch implementation and models for DINOv3
Generative Models by Stability AI
Official PyTorch Implementation of Unified Video Action Model (RSS 2025)
TAPIP3D: Tracking Any Point in Persistent 3D Geometry
[ICCV 2025] SpatialTrackerV2: 3D Point Tracking Made Easy
An open-source, GPU-accelerated physics simulation engine built upon NVIDIA Warp, specifically targeting roboticists and simulation researchers.
Universal Monocular Metric Depth Estimation