Exploring the Coordination of Frequency and Attention in Masked Image Modeling
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910623797870592 |
|---|---|
| author | Gui, Jie Chen, Tuo Dong, Minjing Liu, Zhengqi Luo, Hao Kwok, James Tin-Yau Tang, Yuan Yan |
| author_facet | Gui, Jie Chen, Tuo Dong, Minjing Liu, Zhengqi Luo, Hao Kwok, James Tin-Yau Tang, Yuan Yan |
| contents | Recently, masked image modeling (MIM), which learns visual representations by reconstructing the masked patches of an image, has dominated self-supervised learning in computer vision. However, the pre-training of MIM always takes massive time due to the large-scale data and large-size backbones. We mainly attribute it to the random patch masking in previous MIM works, which fails to leverage the crucial semantic information for effective visual representation learning. To tackle this issue, we propose the Frequency \& Attention-driven Masking and Throwing Strategy (FAMT), which can extract semantic patches and reduce the number of training patches to boost model performance and training efficiency simultaneously. Specifically, FAMT utilizes the self-attention mechanism to extract semantic information from the image for masking during training in an unsupervised manner. However, attention alone could sometimes focus on inappropriate areas regarding the semantic information. Thus, we are motivated to incorporate the information from the frequency domain into the self-attention mechanism to derive the sampling weights for masking, which captures semantic patches for visual representation learning. Furthermore, we introduce a patch throwing strategy based on the derived sampling weights to reduce the training cost. FAMT can be seamlessly integrated as a plug-and-play module and surpasses previous works, \emph{e.g.} reducing the training phase time by nearly $50\%$ and improving the linear probing accuracy of MAE by $1.3\% \sim 3.9\%$ across various datasets, including CIFAR-10/100, Tiny ImageNet, and ImageNet-1K. FAMT also demonstrates superior performance in downstream detection and segmentation tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2211_15362 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Exploring the Coordination of Frequency and Attention in Masked Image Modeling Gui, Jie Chen, Tuo Dong, Minjing Liu, Zhengqi Luo, Hao Kwok, James Tin-Yau Tang, Yuan Yan Computer Vision and Pattern Recognition Machine Learning Recently, masked image modeling (MIM), which learns visual representations by reconstructing the masked patches of an image, has dominated self-supervised learning in computer vision. However, the pre-training of MIM always takes massive time due to the large-scale data and large-size backbones. We mainly attribute it to the random patch masking in previous MIM works, which fails to leverage the crucial semantic information for effective visual representation learning. To tackle this issue, we propose the Frequency \& Attention-driven Masking and Throwing Strategy (FAMT), which can extract semantic patches and reduce the number of training patches to boost model performance and training efficiency simultaneously. Specifically, FAMT utilizes the self-attention mechanism to extract semantic information from the image for masking during training in an unsupervised manner. However, attention alone could sometimes focus on inappropriate areas regarding the semantic information. Thus, we are motivated to incorporate the information from the frequency domain into the self-attention mechanism to derive the sampling weights for masking, which captures semantic patches for visual representation learning. Furthermore, we introduce a patch throwing strategy based on the derived sampling weights to reduce the training cost. FAMT can be seamlessly integrated as a plug-and-play module and surpasses previous works, \emph{e.g.} reducing the training phase time by nearly $50\%$ and improving the linear probing accuracy of MAE by $1.3\% \sim 3.9\%$ across various datasets, including CIFAR-10/100, Tiny ImageNet, and ImageNet-1K. FAMT also demonstrates superior performance in downstream detection and segmentation tasks. |
| title | Exploring the Coordination of Frequency and Attention in Masked Image Modeling |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2211.15362 |