Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2206.12943 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909109709701120 |
|---|---|
| author | Gao, Xiang Tian, Yingjie Qi, Zhiquan |
| author_facet | Gao, Xiang Tian, Yingjie Qi, Zhiquan |
| contents | We propose an end-to-end-trainable feature augmentation module built for image classification that extracts and exploits multi-view local features to boost model performance. Different from using global average pooling (GAP) to extract vectorized features from only the global view, we propose to sample and ensemble diverse multi-view local features to improve model robustness. To sample class-representative local features, we incorporate a simple auxiliary classifier head (comprising only one 1$\times$1 convolutional layer) which efficiently and adaptively attends to class-discriminative local regions of feature maps via our proposed AdaCAM (Adaptive Class Activation Mapping). Extensive experiments demonstrate consistent and noticeable performance gains achieved by our multi-view feature augmentation module. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2206_12943 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Multi-view Feature Augmentation with Adaptive Class Activation Mapping Gao, Xiang Tian, Yingjie Qi, Zhiquan Computer Vision and Pattern Recognition We propose an end-to-end-trainable feature augmentation module built for image classification that extracts and exploits multi-view local features to boost model performance. Different from using global average pooling (GAP) to extract vectorized features from only the global view, we propose to sample and ensemble diverse multi-view local features to improve model robustness. To sample class-representative local features, we incorporate a simple auxiliary classifier head (comprising only one 1$\times$1 convolutional layer) which efficiently and adaptively attends to class-discriminative local regions of feature maps via our proposed AdaCAM (Adaptive Class Activation Mapping). Extensive experiments demonstrate consistent and noticeable performance gains achieved by our multi-view feature augmentation module. |
| title | Multi-view Feature Augmentation with Adaptive Class Activation Mapping |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2206.12943 |