Be Yourself: Bounded Attention for Multi-Subject Text-to-Image Generation
This addresses a key limitation in text-to-image generation for users needing precise multi-subject control, though it is incremental as it builds on existing layout-to-image methods.
The paper tackled the problem of text-to-image diffusion models failing to accurately generate images with multiple subjects, and introduced Bounded Attention, a training-free method that prevents semantic leakage between subjects, resulting in improved alignment with prompts and layouts.
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.