CA-Stream: Attention-based pooling for interpretable image recognition
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917647967322112 |
|---|---|
| author | Torres, Felipe Zhang, Hanwei Sicre, Ronan Ayache, Stéphane Avrithis, Yannis |
| author_facet | Torres, Felipe Zhang, Hanwei Sicre, Ronan Ayache, Stéphane Avrithis, Yannis |
| contents | Explanations obtained from transformer-based architectures in the form of raw attention, can be seen as a class-agnostic saliency map. Additionally, attention-based pooling serves as a form of masking the in feature space. Motivated by this observation, we design an attention-based pooling mechanism intended to replace Global Average Pooling (GAP) at inference. This mechanism, called Cross-Attention Stream (CA-Stream), comprises a stream of cross attention blocks interacting with features at different network depths. CA-Stream enhances interpretability in models, while preserving recognition performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_14996 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CA-Stream: Attention-based pooling for interpretable image recognition Torres, Felipe Zhang, Hanwei Sicre, Ronan Ayache, Stéphane Avrithis, Yannis Computer Vision and Pattern Recognition Explanations obtained from transformer-based architectures in the form of raw attention, can be seen as a class-agnostic saliency map. Additionally, attention-based pooling serves as a form of masking the in feature space. Motivated by this observation, we design an attention-based pooling mechanism intended to replace Global Average Pooling (GAP) at inference. This mechanism, called Cross-Attention Stream (CA-Stream), comprises a stream of cross attention blocks interacting with features at different network depths. CA-Stream enhances interpretability in models, while preserving recognition performance. |
| title | CA-Stream: Attention-based pooling for interpretable image recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.14996 |