Unbiased Image Synthesis via Manifold Guidance in Diffusion Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909170080415744 |
|---|---|
| author | Su, Xingzhe Jia, Daixi Wu, Fengge Zhao, Junsuo Zheng, Changwen Qiang, Wenwen |
| author_facet | Su, Xingzhe Jia, Daixi Wu, Fengge Zhao, Junsuo Zheng, Changwen Qiang, Wenwen |
| contents | Diffusion Models are a potent class of generative models capable of producing high-quality images. However, they often inadvertently favor certain data attributes, undermining the diversity of generated images. This issue is starkly apparent in skewed datasets like CelebA, where the initial dataset disproportionately favors females over males by 57.9%, this bias amplified in generated data where female representation outstrips males by 148%. In response, we propose a plug-and-play method named Manifold Guidance Sampling, which is also the first unsupervised method to mitigate bias issue in DDPMs. Leveraging the inherent structure of the data manifold, this method steers the sampling process towards a more uniform distribution, effectively dispersing the clustering of biased data. Without the need for modifying the existing model or additional training, it significantly mitigates data bias and enhances the quality and unbiasedness of the generated images. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_08199 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Unbiased Image Synthesis via Manifold Guidance in Diffusion Models Su, Xingzhe Jia, Daixi Wu, Fengge Zhao, Junsuo Zheng, Changwen Qiang, Wenwen Computer Vision and Pattern Recognition Diffusion Models are a potent class of generative models capable of producing high-quality images. However, they often inadvertently favor certain data attributes, undermining the diversity of generated images. This issue is starkly apparent in skewed datasets like CelebA, where the initial dataset disproportionately favors females over males by 57.9%, this bias amplified in generated data where female representation outstrips males by 148%. In response, we propose a plug-and-play method named Manifold Guidance Sampling, which is also the first unsupervised method to mitigate bias issue in DDPMs. Leveraging the inherent structure of the data manifold, this method steers the sampling process towards a more uniform distribution, effectively dispersing the clustering of biased data. Without the need for modifying the existing model or additional training, it significantly mitigates data bias and enhances the quality and unbiasedness of the generated images. |
| title | Unbiased Image Synthesis via Manifold Guidance in Diffusion Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2307.08199 |