Distributionally Generative Augmentation for Fair Facial Attribute Classification
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910381232881664 |
|---|---|
| author | Zhang, Fengda He, Qianpei Kuang, Kun Liu, Jiashuo Chen, Long Wu, Chao Xiao, Jun Zhang, Hanwang |
| author_facet | Zhang, Fengda He, Qianpei Kuang, Kun Liu, Jiashuo Chen, Long Wu, Chao Xiao, Jun Zhang, Hanwang |
| contents | Facial Attribute Classification (FAC) holds substantial promise in widespread applications. However, FAC models trained by traditional methodologies can be unfair by exhibiting accuracy inconsistencies across varied data subpopulations. This unfairness is largely attributed to bias in data, where some spurious attributes (e.g., Male) statistically correlate with the target attribute (e.g., Smiling). Most of existing fairness-aware methods rely on the labels of spurious attributes, which may be unavailable in practice. This work proposes a novel, generation-based two-stage framework to train a fair FAC model on biased data without additional annotation. Initially, we identify the potential spurious attributes based on generative models. Notably, it enhances interpretability by explicitly showing the spurious attributes in image space. Following this, for each image, we first edit the spurious attributes with a random degree sampled from a uniform distribution, while keeping target attribute unchanged. Then we train a fair FAC model by fostering model invariance to these augmentation. Extensive experiments on three common datasets demonstrate the effectiveness of our method in promoting fairness in FAC without compromising accuracy. Codes are in https://github.com/heqianpei/DiGA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_06606 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Distributionally Generative Augmentation for Fair Facial Attribute Classification Zhang, Fengda He, Qianpei Kuang, Kun Liu, Jiashuo Chen, Long Wu, Chao Xiao, Jun Zhang, Hanwang Computer Vision and Pattern Recognition Machine Learning Facial Attribute Classification (FAC) holds substantial promise in widespread applications. However, FAC models trained by traditional methodologies can be unfair by exhibiting accuracy inconsistencies across varied data subpopulations. This unfairness is largely attributed to bias in data, where some spurious attributes (e.g., Male) statistically correlate with the target attribute (e.g., Smiling). Most of existing fairness-aware methods rely on the labels of spurious attributes, which may be unavailable in practice. This work proposes a novel, generation-based two-stage framework to train a fair FAC model on biased data without additional annotation. Initially, we identify the potential spurious attributes based on generative models. Notably, it enhances interpretability by explicitly showing the spurious attributes in image space. Following this, for each image, we first edit the spurious attributes with a random degree sampled from a uniform distribution, while keeping target attribute unchanged. Then we train a fair FAC model by fostering model invariance to these augmentation. Extensive experiments on three common datasets demonstrate the effectiveness of our method in promoting fairness in FAC without compromising accuracy. Codes are in https://github.com/heqianpei/DiGA. |
| title | Distributionally Generative Augmentation for Fair Facial Attribute Classification |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2403.06606 |