franciszzj/Leffa
A diffusion-based image generation framework for virtual try-on and pose transfer using learned flow fields within attention layers.
Not currently ranked — collecting fresh signals.
star history
Leffa is a unified framework for controllable person image generation that enables precise manipulation of both appearance (virtual try-on) and pose (pose transfer). It learns flow fields within the attention mechanism of diffusion models to achieve better warping and feature alignment between reference and target poses. The project provides pre-trained models on DeepFashion and DressCode datasets, a HuggingFace demo, and has been integrated into ComfyUI for workflow use.
Frequently asked
- What is franciszzj/Leffa?
- A diffusion-based image generation framework for virtual try-on and pose transfer using learned flow fields within attention layers.
- Is Leffa open source?
- Yes — franciszzj/Leffa is open source, released under the MIT license.
- What language is Leffa written in?
- franciszzj/Leffa is primarily written in Python.
- How popular is Leffa?
- franciszzj/Leffa has 1.7k stars on GitHub.
- Where can I find Leffa?
- franciszzj/Leffa is on GitHub at https://github.com/franciszzj/Leffa.