@article{oai:ipsj.ixsq.nii.ac.jp:00231556,
 author = {Hitoshi, Nakanishi and Masahiro, Suzuki and Yutaka, Matsuo and Hitoshi, Nakanishi and Masahiro, Suzuki and Yutaka, Matsuo},
 issue = {12},
 journal = {情報処理学会論文誌},
 month = {Dec},
 note = {To handle high-resolution images on finite computational resources, many researches have been conducted on hierarchical networks to load features in only the most meaningful local regions. However, it is difficult to determine the correct number and location of patch regions at the appropriate scale in these methods. Then, incorrectly selected regions at different scales interfere with feature extraction and information integration. To solve this issue, we propose a hierarchical attention weighted network (HAWK-Net), which consists of a backbone network with differentiable Top-K selection and spatially gated blocks. The Top-K selected patches are identified from multiple image scaled features and extracted from an original high-resolution image. Then, patch features are aggregated via a novel gate mechanism under the uncertainty of the predicted information. Not only can multi-scale information uncertainty be modeled, but it also controls the gradient to the feature network coming from patch images with low confidence in the region proposal network in feedback during training. Our model is a simple yet efficient network structure that can learn from multiple scales and patches and is capable of end-to-end training. Based on benchmarks of multiple high-resolution images, our model achieves even higher performance with lower memory usage and reduced computation time.
------------------------------
This is a preprint of an article intended for publication Journal of
Information Processing(JIP). This preprint should not be cited. This
article should be cited as: Journal of Information Processing Vol.31(2023) (online)
DOI　http://dx.doi.org/10.2197/ipsjjip.31.851
------------------------------, To handle high-resolution images on finite computational resources, many researches have been conducted on hierarchical networks to load features in only the most meaningful local regions. However, it is difficult to determine the correct number and location of patch regions at the appropriate scale in these methods. Then, incorrectly selected regions at different scales interfere with feature extraction and information integration. To solve this issue, we propose a hierarchical attention weighted network (HAWK-Net), which consists of a backbone network with differentiable Top-K selection and spatially gated blocks. The Top-K selected patches are identified from multiple image scaled features and extracted from an original high-resolution image. Then, patch features are aggregated via a novel gate mechanism under the uncertainty of the predicted information. Not only can multi-scale information uncertainty be modeled, but it also controls the gradient to the feature network coming from patch images with low confidence in the region proposal network in feedback during training. Our model is a simple yet efficient network structure that can learn from multiple scales and patches and is capable of end-to-end training. Based on benchmarks of multiple high-resolution images, our model achieves even higher performance with lower memory usage and reduced computation time.
------------------------------
This is a preprint of an article intended for publication Journal of
Information Processing(JIP). This preprint should not be cited. This
article should be cited as: Journal of Information Processing Vol.31(2023) (online)
DOI　http://dx.doi.org/10.2197/ipsjjip.31.851
------------------------------},
 title = {HAWK-Net: Hierarchical Attention Weighted Top-K Network for High-resolution Image Classification},
 volume = {64},
 year = {2023}
}