StorySync: Training-Free Subject Consistency in Text-to-Image Generation via Region Harmonization
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911093569355776 |
|---|---|
| author | Gaur, Gopalji Zolfaghari, Mohammadreza Brox, Thomas |
| author_facet | Gaur, Gopalji Zolfaghari, Mohammadreza Brox, Thomas |
| contents | Generating a coherent sequence of images that tells a visual story, using text-to-image diffusion models, often faces the critical challenge of maintaining subject consistency across all story scenes. Existing approaches, which typically rely on fine-tuning or retraining models, are computationally expensive, time-consuming, and often interfere with the model's pre-existing capabilities. In this paper, we follow a training-free approach and propose an efficient consistent-subject-generation method. This approach works seamlessly with pre-trained diffusion models by introducing masked cross-image attention sharing to dynamically align subject features across a batch of images, and Regional Feature Harmonization to refine visually similar details for improved subject consistency. Experimental results demonstrate that our approach successfully generates visually consistent subjects across a variety of scenarios while maintaining the creative abilities of the diffusion model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_03735 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | StorySync: Training-Free Subject Consistency in Text-to-Image Generation via Region Harmonization Gaur, Gopalji Zolfaghari, Mohammadreza Brox, Thomas Computer Vision and Pattern Recognition Artificial Intelligence Generating a coherent sequence of images that tells a visual story, using text-to-image diffusion models, often faces the critical challenge of maintaining subject consistency across all story scenes. Existing approaches, which typically rely on fine-tuning or retraining models, are computationally expensive, time-consuming, and often interfere with the model's pre-existing capabilities. In this paper, we follow a training-free approach and propose an efficient consistent-subject-generation method. This approach works seamlessly with pre-trained diffusion models by introducing masked cross-image attention sharing to dynamically align subject features across a batch of images, and Regional Feature Harmonization to refine visually similar details for improved subject consistency. Experimental results demonstrate that our approach successfully generates visually consistent subjects across a variety of scenarios while maintaining the creative abilities of the diffusion model. |
| title | StorySync: Training-Free Subject Consistency in Text-to-Image Generation via Region Harmonization |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2508.03735 |