Be Yourself: Bounded Attention for Multi-Subject Text-to-Image Generation
Authors:
Preprint
Abstract
Text-to-image diffusion models have an unprecedented ability to generate diverse and high-quality images. However, they often struggle to faithfully capture the intended semantics of complex input prompts that include multiple subjects. Recently, numerous layout-to-image extensions have been introduced to improve user control, aiming to localize subjects represented by specific tokens. Yet, these methods often produce semantically inaccurate images, especially when dealing with multiple semantically or visually similar subjects. In this work, we study and analyze the causes of these limitations. Our exploration reveals that the primary issue stems from inadvertent semantic leakage between subjects in the denoising process. This leakage is attributed to the diffusion model's attention layers, which tend to blend the visual features of different subjects. To address these issues, we introduce Bounded Attention, a training-free method for bounding the information flow in the sampling process. Bounded Attention prevents detrimental leakage among subjects and enables guiding the generation to promote each subject's individuality, even with complex multi-subject conditioning. Through extensive experimentation, we demonstrate that our method empowers the generation of multiple subjects that better align with given prompts and layouts.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Grounded Text-to-Image Synthesis with Attention Refocusing2024
- Training-Free Layout Control with Cross-Attention Guidance2024
- Masked-attention diffusion guidance for spatially controlling text-to-image generationThe Visual Computer, 2023
- LayoutLLM-T2I: Eliciting Layout Guidance from LLM for Text-to-Image Generation2023
- MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing2023
- Teaching CLIP to Count to Ten2023
- Localizing Object-level Shape Variations with Text-to-Image Diffusion Models2023
- Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion ModelsACM Transactions on Graphics, 2023