APPLE: Attribute-Preserving Pseudo-Labeling for Diffusion-Based Face Swapping
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915900006858752 |
|---|---|
| author | Kang, Jiwon Choi, Yeji Lee, JoungBin Jang, Wooseok Choi, Jinhyeok Kang, Taekeun Park, Yongjae Kim, Myungin Kim, Seungryong |
| author_facet | Kang, Jiwon Choi, Yeji Lee, JoungBin Jang, Wooseok Choi, Jinhyeok Kang, Taekeun Park, Yongjae Kim, Myungin Kim, Seungryong |
| contents | Face swapping aims to transfer the identity of a source face onto a target face while preserving target-specific attributes such as pose, expression, lighting, skin tone, and makeup. However, since real ground truth for face swapping is unavailable, achieving both accurate identity transfer and high-quality attribute preservation remains challenging. Recent diffusion-based approaches attempt to improve visual fidelity through conditional inpainting on masked target images, but the masked condition removes crucial appearance cues, resulting in plausible yet misaligned attributes. To address this limitation, we propose APPLE (Attribute-Preserving Pseudo-Labeling), a fully diffusion-based teacher-student framework for attribute-preserving face swapping. Our approach introduces a teacher design to produce pseudo-labels aligned with the target attributes through (1) a conditional deblurring formulation that improves the preservation of global attributes such as skin tone and illumination, and (2) an attribute-aware inversion scheme that further enhances fine-grained attribute preservation such as makeup. APPLE conditions the student on clean pseudo-labels rather than degraded masked inputs, enabling more faithful attribute preservation. As a result, APPLE achieves state-of-the-art performance in attribute preservation while maintaining competitive identity transferability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15288 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | APPLE: Attribute-Preserving Pseudo-Labeling for Diffusion-Based Face Swapping Kang, Jiwon Choi, Yeji Lee, JoungBin Jang, Wooseok Choi, Jinhyeok Kang, Taekeun Park, Yongjae Kim, Myungin Kim, Seungryong Computer Vision and Pattern Recognition Face swapping aims to transfer the identity of a source face onto a target face while preserving target-specific attributes such as pose, expression, lighting, skin tone, and makeup. However, since real ground truth for face swapping is unavailable, achieving both accurate identity transfer and high-quality attribute preservation remains challenging. Recent diffusion-based approaches attempt to improve visual fidelity through conditional inpainting on masked target images, but the masked condition removes crucial appearance cues, resulting in plausible yet misaligned attributes. To address this limitation, we propose APPLE (Attribute-Preserving Pseudo-Labeling), a fully diffusion-based teacher-student framework for attribute-preserving face swapping. Our approach introduces a teacher design to produce pseudo-labels aligned with the target attributes through (1) a conditional deblurring formulation that improves the preservation of global attributes such as skin tone and illumination, and (2) an attribute-aware inversion scheme that further enhances fine-grained attribute preservation such as makeup. APPLE conditions the student on clean pseudo-labels rather than degraded masked inputs, enabling more faithful attribute preservation. As a result, APPLE achieves state-of-the-art performance in attribute preservation while maintaining competitive identity transferability. |
| title | APPLE: Attribute-Preserving Pseudo-Labeling for Diffusion-Based Face Swapping |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.15288 |