From Parts to Whole: A Unified Reference Framework for Controllable Human Image Generation
Paper • 2404.15267 • Published • 5
Model Card for Parts2Whole
Parts2Whole, a novel framework designed for generating customized portraits from multiple reference images, including pose images and various aspects of human appearance.
Model Details
Model Description
- Developed by: Zehuan Huang, Hongxing Fan, Lipeng Wang
- Model type: human image diffusion model
- License: apache-2.0
- Finetuned from model [optional]: Stable Diffusion 1-5
Model Sources [optional]
- Repository: https://github.com/huanngzh/Parts2Whole
- Paper [optional]: https://arxiv.org/abs/2404.15267
- Demo [optional]: [More Information Needed]
Uses
For usage instructions, please refer to our Parts2Whole GitHub repository.
Training Details
Training Data
https://huggingface.co/datasets/huanngzh/DeepFashion-Refs2Target
- Downloads last month
- 5
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support