UE5-based Data Engine used in OmniX

Official implementation paper OmniX: Any-view and Any-time 4D Reconstruction via Feed-forward Trajectory Fields

[CVPR 2025 Highlight] Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

🔥 Data Pyramid for Embodied Manipulation: A Survey

Official implementation of LaMP: Learning Vision-Language-Action Policy with 3D Scene Flow as Latent Motion Prior.

Code for SM4RT: Learning Structured Motion Geometry for 4D Reconstruction

RynnWorld-4D: 4D Embodied World Models for Robotic Manipulation

Infinite Worlds with Versatile Interactions

Code for MIRA: Multiplayer Interactive World Models with Representation Autoencoders

A high-performance Python-based I/O system for large (and small) deep learning problems, with strong support for PyTorch.

Official implementation of "Repurposing Geometric Foundation Models for Multi-view Diffusion"

Lifting Monocular Video for 4D Multi-Object Interactions via Human-Agent Collaboration

Official implementation of "Geometric Action Model for Robot Policy Learning"

[ECCV 2026] Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels

[CVPR 2026 Oral] Official implementation for "S²AM3D: Scale-controllable Part Segmentation of 3D Point Clouds"

RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI

HorizonStream Official Implementation

A Web-Scale 4D Hand-Object Interaction Data Engine for Any-View Robot Retargeting and Video-to-Action Robot Learning

Prepreint

Code for CVPR2026 Paper "Revisiting Articulated Parts Perception in Robot Manipulation"

[ICML 2026] Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation" & Causal Forcing++

Web-based 3D visualization in Python

Reimplementation of LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory

Official code for "LagerNVS Latent Geometry for Fully Neural Real-time Novel View Synthesis" (CVPR 2026)