Tencent/MimicMotion
High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance
Leverages confidence-weighted pose estimation (DWPose) to guide diffusion-based video generation while mitigating artifacts through regional loss amplification, and implements progressive latent fusion to synthesize arbitrary-length videos (up to 72 frames) efficiently. Built on Stable Video Diffusion with a custom U-Net architecture, it targets PyTorch 2.x environments and integrates Hugging Face model checkpoints for streamlined deployment.
2,546 stars.
Stars
2,546
Forks
227
Language
Python
License
—
Category
Last pushed
Nov 18, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Tencent/MimicMotion"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
hao-ai-lab/FastVideo
A unified inference and post-training framework for accelerated video generation.
thu-ml/TurboDiffusion
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
ModelTC/LightX2V
Light Image Video Generation Inference Framework
PKU-YuanGroup/Helios
Helios: Real Real-Time Long Video Generation Model
PKU-YuanGroup/MagicTime
[TPAMI 2025🔥] MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators