Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2409.06963 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918058323345408 |
|---|---|
| author | Yu, Yonghao Zhao, Dongcheng Shen, Guobin Dong, Yiting Zeng, Yi |
| author_facet | Yu, Yonghao Zhao, Dongcheng Shen, Guobin Dong, Yiting Zeng, Yi |
| contents | The hierarchical architecture has become a mainstream design paradigm for Vision Transformers (ViTs), with Patch Merging serving as the pivotal component that transforms a columnar architecture into a hierarchical one. Drawing inspiration from the brain's ability to integrate global and local information for comprehensive visual understanding, we propose Stepwise Patch Merging (SPM), which enhances the subsequent attention mechanism's ability to 'see' better. SPM consists of Multi-Scale Aggregation (MSA) and Guided Local Enhancement (GLE) striking a proper balance between long-range dependency modeling and local feature enhancement. Extensive experiments conducted on benchmark datasets, including ImageNet-1K, COCO, and ADE20K, demonstrate that SPM significantly improves the performance of various models, particularly in dense prediction tasks such as object detection and semantic segmentation. Meanwhile, experiments show that combining SPM with different backbones can further improve performance. The code has been released at https://github.com/Yonghao-Yu/StepwisePatchMerging. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_06963 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Brain-Inspired Stepwise Patch Merging for Vision Transformers Yu, Yonghao Zhao, Dongcheng Shen, Guobin Dong, Yiting Zeng, Yi Computer Vision and Pattern Recognition The hierarchical architecture has become a mainstream design paradigm for Vision Transformers (ViTs), with Patch Merging serving as the pivotal component that transforms a columnar architecture into a hierarchical one. Drawing inspiration from the brain's ability to integrate global and local information for comprehensive visual understanding, we propose Stepwise Patch Merging (SPM), which enhances the subsequent attention mechanism's ability to 'see' better. SPM consists of Multi-Scale Aggregation (MSA) and Guided Local Enhancement (GLE) striking a proper balance between long-range dependency modeling and local feature enhancement. Extensive experiments conducted on benchmark datasets, including ImageNet-1K, COCO, and ADE20K, demonstrate that SPM significantly improves the performance of various models, particularly in dense prediction tasks such as object detection and semantic segmentation. Meanwhile, experiments show that combining SPM with different backbones can further improve performance. The code has been released at https://github.com/Yonghao-Yu/StepwisePatchMerging. |
| title | Brain-Inspired Stepwise Patch Merging for Vision Transformers |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.06963 |