We are the HumanAIGC Team at Tongyi, Alibaba. We focus on the understanding and generation of human-centric content.
- Exploring Timeline Control for Facial Motion Generation. Project, Paper
- MotionRAG-Diff: A Retrieval-Augmented Diffusion Framework for Long-Term Music-to-Dance Generation. Project, Paper
- OmniTalker: Real-Time Text-Driven Talking Head Generation with In-Context Audio-Visual Style Replication. Project, Paper, Demo
- ChatAnyone: Stylized Real-time Portrait Video Generation with Hierarchical Motion Diffusion Model. Project, Paper
- LiteAvatar: a audio2face model for realtime 2D chat avatar. Code
- Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance. Project, Paper
- EMO2: End-Effector Guided Audio-Driven Avatar Video Generation. Project, Paper
- EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions. Project, Paper
- Outfit Anyone: Ultra-high quality virtual try-on for Any Clothing and Any Person. Project, Paper, Demo
- Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation. Project, Paper
- VividTalk: One-Shot Audio-Driven Talking Head Generation Based on 3D Hybrid Prior. Project, Paper
- DanceMeld: Unraveling Dance Phrases with Hierarchical Latent Codes for Music-to-Dance Synthesis. Paper
- Cloth2Tex: A Customized Cloth Texture Generation Pipeline for 3D Virtual Try-On. Project, Paper, Code