AttenCraft: Attention-guided Disentanglement of Multiple Concepts for Text-to-Image Customization

The dataset used in the paper is a collection of images with multiple concepts, where the goal is to disentangle multiple concepts from a single image.

Data and Resources

Cite this as

Junjie Shentu, Matthew Watson, Noura Al Moubayed (2024). Dataset: AttenCraft: Attention-guided Disentanglement of Multiple Concepts for Text-to-Image Customization. https://doi.org/10.57702/kxml2p4q

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2405.17965
Author Junjie Shentu
More Authors
Matthew Watson
Noura Al Moubayed
Homepage https://github.com/junjie-shentu/AttenCraft