MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance

TL;DR

MS-Diffusion is a novel framework that enables zero-shot, layout-guided multi-subject image personalization in text-to-image generation, maintaining subject details and ensuring cohesive multi-subject compositions.

Contribution

The paper introduces MS-Diffusion, a new approach integrating grounding tokens and enhanced cross-attention for improved multi-subject image personalization with layout guidance.

Findings

Outperforms existing models in image fidelity.

Achieves better preservation of subject details.

Ensures cohesive multi-subject compositions.

Abstract

Recent advancements in text-to-image generation models have dramatically enhanced the generation of photorealistic images from textual prompts, leading to an increased interest in personalized text-to-image applications, particularly in multi-subject scenarios. However, these advances are hindered by two main challenges: firstly, the need to accurately maintain the details of each referenced subject in accordance with the textual descriptions; and secondly, the difficulty in achieving a cohesive representation of multiple subjects in a single image without introducing inconsistencies. To address these concerns, our research introduces the MS-Diffusion framework for layout-guided zero-shot image personalization with multi-subjects. This innovative approach integrates grounding tokens with the feature resampler to maintain detail fidelity among subjects. With the layout guidance, MS-Diffusion further improves the cross-attention to adapt to the multi-subject inputs, ensuring that each subject condition acts on specific areas. The proposed multi-subject cross-attention orchestrates harmonious inter-subject compositions while preserving the control of texts. Comprehensive quantitative and qualitative experiments affirm that this method surpasses existing models in both image and text fidelity, promoting the development of personalized text-to-image generation. The project page is https://MS-Diffusion.github.io.