DisenStudio: Customized Multi-subject Text-to-Video Generation with Disentangled Spatial Control

TL;DR

DisenStudio is a novel framework that enables multi-subject, text-guided video generation with disentangled spatial control, addressing previous issues of subject-missing, attribute-binding, and action-binding in multi-subject scenarios.

Contribution

The paper introduces DisenStudio, a diffusion-based model with spatial-disentangled cross-attention and specialized finetuning strategies for effective multi-subject text-to-video generation.

Findings

Outperforms existing methods in multiple metrics.

Successfully generates videos with multiple subjects and desired actions.

Demonstrates versatility in controllable generation applications.

Abstract

Generating customized content in videos has received increasing attention recently. However, existing works primarily focus on customized text-to-video generation for single subject, suffering from subject-missing and attribute-binding problems when the video is expected to contain multiple subjects. Furthermore, existing models struggle to assign the desired actions to the corresponding subjects (action-binding problem), failing to achieve satisfactory multi-subject generation performance. To tackle the problems, in this paper, we propose DisenStudio, a novel framework that can generate text-guided videos for customized multiple subjects, given few images for each subject. Specifically, DisenStudio enhances a pretrained diffusion-based text-to-video model with our proposed spatial-disentangled cross-attention mechanism to associate each subject with the desired action. Then the model is customized for the multiple subjects with the proposed motion-preserved disentangled finetuning, which involves three tuning strategies: multi-subject co-occurrence tuning, masked single-subject tuning, and multi-subject motion-preserved tuning. The first two strategies guarantee the subject occurrence and preserve their visual attributes, and the third strategy helps the model maintain the temporal motion-generation ability when finetuning on static images. We conduct extensive experiments to demonstrate our proposed DisenStudio significantly outperforms existing methods in various metrics. Additionally, we show that DisenStudio can be used as a powerful tool for various controllable generation applications.