PuLID: Pure and Lightning ID Customization via Contrastive Alignment
This work addresses the challenge of customizing identities in AI-generated images for applications in digital art and content creation, representing an incremental improvement over existing methods.
The paper tackles the problem of ID customization in text-to-image generation by proposing PuLID, a tuning-free method that achieves high ID fidelity and editability while maintaining consistency in image elements like background and style.
We propose Pure and Lightning ID customization (PuLID), a novel tuning-free ID customization method for text-to-image generation. By incorporating a Lightning T2I branch with a standard diffusion one, PuLID introduces both contrastive alignment loss and accurate ID loss, minimizing disruption to the original model and ensuring high ID fidelity. Experiments show that PuLID achieves superior performance in both ID fidelity and editability. Another attractive property of PuLID is that the image elements (e.g., background, lighting, composition, and style) before and after the ID insertion are kept as consistent as possible. Codes and models are available at https://github.com/ToTheBeginning/PuLID