Geometric Disentanglement of Text Embeddings for Subject-Consistent Text-to-Image Generation using A Single Prompt
This work addresses a key limitation in text-to-image generation for applications like visual storytelling, offering a more efficient solution compared to fine-tuning or image conditioning methods.
The paper tackles the problem of subject inconsistency in text-to-image diffusion models by proposing a training-free method that refines text embeddings to suppress unwanted semantics, resulting in significant improvements in both subject consistency and text alignment over existing baselines.
Text-to-image diffusion models excel at generating high-quality images from natural language descriptions but often fail to preserve subject consistency across multiple outputs, limiting their use in visual storytelling. Existing approaches rely on model fine-tuning or image conditioning, which are computationally expensive and require per-subject optimization. 1Prompt1Story, a training-free approach, concatenates all scene descriptions into a single prompt and rescales token embeddings, but it suffers from semantic leakage, where embeddings across frames become entangled, causing text misalignment. In this paper, we propose a simple yet effective training-free approach that addresses semantic entanglement from a geometric perspective by refining text embeddings to suppress unwanted semantics. Extensive experiments prove that our approach significantly improves both subject consistency and text alignment over existing baselines.