Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2305.19599 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929555910950912 |
|---|---|
| author | Jiang, Zutao Fang, Guian Han, Jianhua Lu, Guansong Xu, Hang Liao, Shengcai Chang, Xiaojun Liang, Xiaodan |
| author_facet | Jiang, Zutao Fang, Guian Han, Jianhua Lu, Guansong Xu, Hang Liao, Shengcai Chang, Xiaojun Liang, Xiaodan |
| contents | Recent advances in text-to-image diffusion models have achieved remarkable success in generating high-quality, realistic images from textual descriptions. However, these approaches have faced challenges in precisely aligning the generated visual content with the textual concepts described in the prompts. In this paper, we propose a two-stage coarse-to-fine semantic re-alignment method, named RealignDiff, aimed at improving the alignment between text and images in text-to-image diffusion models. In the coarse semantic re-alignment phase, a novel caption reward, leveraging the BLIP-2 model, is proposed to evaluate the semantic discrepancy between the generated image caption and the given text prompt. Subsequently, the fine semantic re-alignment stage employs a local dense caption generation module and a re-weighting attention modulation module to refine the previously generated images from a local semantic view. Experimental results on the MS-COCO and ViLG-300 datasets demonstrate that the proposed two-stage coarse-to-fine semantic re-alignment method outperforms other baseline re-alignment techniques by a substantial margin in both visual quality and semantic similarity with the input prompt. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2305_19599 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | RealignDiff: Boosting Text-to-Image Diffusion Model with Coarse-to-fine Semantic Re-alignment Jiang, Zutao Fang, Guian Han, Jianhua Lu, Guansong Xu, Hang Liao, Shengcai Chang, Xiaojun Liang, Xiaodan Computer Vision and Pattern Recognition Artificial Intelligence Recent advances in text-to-image diffusion models have achieved remarkable success in generating high-quality, realistic images from textual descriptions. However, these approaches have faced challenges in precisely aligning the generated visual content with the textual concepts described in the prompts. In this paper, we propose a two-stage coarse-to-fine semantic re-alignment method, named RealignDiff, aimed at improving the alignment between text and images in text-to-image diffusion models. In the coarse semantic re-alignment phase, a novel caption reward, leveraging the BLIP-2 model, is proposed to evaluate the semantic discrepancy between the generated image caption and the given text prompt. Subsequently, the fine semantic re-alignment stage employs a local dense caption generation module and a re-weighting attention modulation module to refine the previously generated images from a local semantic view. Experimental results on the MS-COCO and ViLG-300 datasets demonstrate that the proposed two-stage coarse-to-fine semantic re-alignment method outperforms other baseline re-alignment techniques by a substantial margin in both visual quality and semantic similarity with the input prompt. |
| title | RealignDiff: Boosting Text-to-Image Diffusion Model with Coarse-to-fine Semantic Re-alignment |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2305.19599 |