Identity-Preserving-Virtual-TryOn
What it is
A LoRA-based controlled diffusion framework designed for personalized virtual try-on. It aims to jointly preserve subject identity and garment fidelity (texture, patterns, and logos) while allowing for prompt-controlled generation.
The problem it solves
Existing methods often suffer from identity drift or lack of fine-grained garment attribute preservation during the try-on process.
What you could use it for
- Personalized virtual try-on for e-commerce
- Generating images of specific individuals wearing different garments
- Maintaining fine-grained garment details like logos and patterns during image generation
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.