muzishen/IMAGPose

[NeurIPS 2024] 🕺IMAGPose🕺: A Unified Conditional Framework for Pose-Guided Person Generation. IMAGPose enables versatile pose-guided image generation with high detail fidelity, pose alignment, and cross-view consistency, overcoming limitations of existing methods.

37
/ 100
Emerging

This project helps create realistic images of people in various poses. You provide an existing image of a person and a target pose (or multiple poses), and it generates new images of that person adopting the desired stance, even from different camera angles. This is ideal for content creators, marketers, and designers who need to visualize a person in different body positions without re-shooting.

349 stars. No commits in the last 6 months.

Use this if you need to generate high-quality images of a person in new poses, or apply multiple poses simultaneously to a single source image, or use multi-view source images for improved realism.

Not ideal if you're looking to generate images of objects or scenes, or if you don't have a specific source person image as input.

virtual-photography digital-fashion character-design content-creation visual-marketing
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

349

Forks

11

Language

Python

License

Apache-2.0

Last pushed

Sep 30, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/muzishen/IMAGPose"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.