PerFace: Metric Learning in Perceptual Facial Similarity for Enhanced Face Anonymization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912602901184512 |
|---|---|
| author | Kumagai, Haruka Wöhler, Leslie Ikehata, Satoshi Aizawa, Kiyoharu |
| author_facet | Kumagai, Haruka Wöhler, Leslie Ikehata, Satoshi Aizawa, Kiyoharu |
| contents | In response to rising societal awareness of privacy concerns, face anonymization techniques have advanced, including the emergence of face-swapping methods that replace one identity with another. Achieving a balance between anonymity and naturalness in face swapping requires careful selection of identities: overly similar faces compromise anonymity, while dissimilar ones reduce naturalness. Existing models, however, focus on binary identity classification "the same person or not", making it difficult to measure nuanced similarities such as "completely different" versus "highly similar but different." This paper proposes a human-perception-based face similarity metric, creating a dataset of 6,400 triplet annotations and metric learning to predict the similarity. Experimental results demonstrate significant improvements in both face similarity prediction and attribute-based face classification tasks over existing methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_20281 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PerFace: Metric Learning in Perceptual Facial Similarity for Enhanced Face Anonymization Kumagai, Haruka Wöhler, Leslie Ikehata, Satoshi Aizawa, Kiyoharu Computer Vision and Pattern Recognition In response to rising societal awareness of privacy concerns, face anonymization techniques have advanced, including the emergence of face-swapping methods that replace one identity with another. Achieving a balance between anonymity and naturalness in face swapping requires careful selection of identities: overly similar faces compromise anonymity, while dissimilar ones reduce naturalness. Existing models, however, focus on binary identity classification "the same person or not", making it difficult to measure nuanced similarities such as "completely different" versus "highly similar but different." This paper proposes a human-perception-based face similarity metric, creating a dataset of 6,400 triplet annotations and metric learning to predict the similarity. Experimental results demonstrate significant improvements in both face similarity prediction and attribute-based face classification tasks over existing methods. |
| title | PerFace: Metric Learning in Perceptual Facial Similarity for Enhanced Face Anonymization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.20281 |