Data Augmentation Through Random Style Replacement
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909807764570112 |
|---|---|
| author | Yang, Qikai Ji, Cheng Luo, Huaiying Li, Panfeng Ding, Zhicheng |
| author_facet | Yang, Qikai Ji, Cheng Luo, Huaiying Li, Panfeng Ding, Zhicheng |
| contents | In this paper, we introduce a novel data augmentation technique that combines the advantages of style augmentation and random erasing by selectively replacing image subregions with style-transferred patches. Our approach first applies a random style transfer to training images, then randomly substitutes selected areas of these images with patches derived from the style-transferred versions. This method is able to seamlessly accommodate a wide range of existing style transfer algorithms and can be readily integrated into diverse data augmentation pipelines. By incorporating our strategy, the training process becomes more robust and less prone to overfitting. Comparative experiments demonstrate that, relative to previous style augmentation methods, our technique achieves superior performance and faster convergence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_10563 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Data Augmentation Through Random Style Replacement Yang, Qikai Ji, Cheng Luo, Huaiying Li, Panfeng Ding, Zhicheng Computer Vision and Pattern Recognition In this paper, we introduce a novel data augmentation technique that combines the advantages of style augmentation and random erasing by selectively replacing image subregions with style-transferred patches. Our approach first applies a random style transfer to training images, then randomly substitutes selected areas of these images with patches derived from the style-transferred versions. This method is able to seamlessly accommodate a wide range of existing style transfer algorithms and can be readily integrated into diverse data augmentation pipelines. By incorporating our strategy, the training process becomes more robust and less prone to overfitting. Comparative experiments demonstrate that, relative to previous style augmentation methods, our technique achieves superior performance and faster convergence. |
| title | Data Augmentation Through Random Style Replacement |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.10563 |