-
GairLab
- Shenzhen
- https://cckaixin.github.io/myWebsite/
Highlights
- Pro
Lists (15)
Sort Name ascending (A-Z)
Stars
AISystem 主要是指AI系统,包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术
AIInfra(AI 基础设施)指AI系统从底层芯片等硬件,到上层软件栈支持AI大模型训练和推理。
RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation
Official PyTorch implementation for "Large Language Diffusion Models"
Code for "Closing the Train-Test Gap in World Models for Gradient-Based Planning"
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.
Learning Deep Representations of Data Distributions
Code of "MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulation"
The first behavioral foundation model to control a virtual physics-based humanoid agent for a wide range of whole-body tasks.
The official codebase of paper "GMT: General Motion Tracking for Humanoid Whole-Body Control"
[arXiv 2025] TWIST2: Scalable, Portable, and Holistic Humanoid Data Collection System
SLOPER4D: A Scene-Aware Dataset for Global 4D Human Pose Estimation in Urban Environments (CVPR2023)
GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware.
Phase-Functioned Neural Networks for Character Control
A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.
[NeurIPS 2024 Spotlight] CooHOI: Learning Cooperative Human-Object Interaction with Manipulated Object Dynamics
Official implementation for "Generating Diverse and Natural 3D Human Motions from Texts (CVPR2022)."
Code for "CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects", NeurIPS 2025
A list of Human-Object Interaction Learning.
Vector (and Scalar) Quantization, in Pytorch
The official implementation of Autoregressive Image Generation using Residual Quantization (CVPR '22)
Official implementation of "MoMask: Generative Masked Modeling of 3D Human Motions (CVPR2024)"
HumanML3D: A large and diverse 3d human motion-language dataset.
Official implementation of OpenTrack.