FlashMLA: Efficient MLA decoding kernels

[NeurIPS 2024] CV-VAE: A Compatible Video VAE for Latent Generative Video Models

I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models

3D Reconstruction for all

[CVPR 2025] Official implementation of "MangaNinja: Line Art Colorization with Precise Reference Following"

Text-to-3D & Image-to-3D & Mesh Exportation with NeRF + Diffusion.

SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images

A Large-scale High-Quality Synthetic Facial depth Dataset and Detailed deep learning-based monocular depth estimation from a single input image.

[NeurIPS 2024] Generalizable and Animatable Gaussian Head Avatar

[ICLR 2024] Generalizable and Precise Head Avatar from Image(s)

Python tools for 3D face: 3DMM, Mesh processing(transform, camera, light, render), 3D face representations.

FaceScape (PAMI2023 & CVPR2020)

[NeurIPS 2024] Generalizable and Animatable Gaussian Head Avatar

A generative world for general-purpose robotics & embodied AI learning.

Official code for paper: Synergizing Motion and Appearance: Multi-Scale Compensatory Codebooks for Talking Head Video Generation

[NeurIPS 2024 Best Paper][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ult…

[CVPR 2024] TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding

[CVPR'24] Interactive3D: Create What You Want by Interactive 3D Generation

[CVPR 2025] EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation

Stable Video Diffusion Training Code and Extensions.

VideoSys: An easy and efficient system for video generation

text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

[ACL 2024 🔥] Video-ChatGPT is a video conversation model capable of generating meaningful conversation about videos. It combines the capabilities of LLMs with a pretrained visual encoder adapted fo…

一个超轻量级、可以在移动端实时运行的数字人模型

High-resolution models for human tasks.

Official code for "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching"

Netflix-level subtitle cutting, translation, alignment, and even dubbing - one-click fully automated AI video subtitle team | Netflix级字幕切割、翻译、对齐、甚至加上配音，一键全自动视频搬运AI字幕组

Hallo2: Long-Duration and High-Resolution Audio-driven Portrait Image Animation