フルテキストファイル
著者
西山 正志 Graduate School of Engineering, Tottori University 研究者総覧 KAKEN
Inoue, Michiko Graduate School of Engineering, Tottori University
岩井 儀雄 Graduate School of Engineering, Tottori University 研究者総覧 KAKEN
キーワード
gaze distribution
attention mechanism
convolutional neural network
gender recognition
self-attention
抄録
We propose an attention mechanism in deep learning networks for gender recognition using the gaze distribution of human observers when they judge the gender of people in pedestrian images. Prevalent attention mechanisms spatially compute the correlation among values of all cells in an input feature map to calculate attention weights. If a large bias in the background of pedestrian images (e.g., test samples and training samples containing different backgrounds) is present, the attention weights learned using the prevalent attention mechanisms are affected by the bias, which in turn reduces the accuracy of gender recognition. To avoid this problem, we incorporate an attention mechanism called gaze-guided self-attention (GSA) that is inspired by human visual attention. Our method assigns spatially suitable attention weights to each input feature map using the gaze distribution of human observers. In particular, GSA yields promising results even when using training samples with the background bias. The results of experiments on publicly available datasets confirm that our GSA, using the gaze distribution, is more accurate in gender recognition than currently available attention-based methods in the case of background bias between training and test samples.
出版者
IEICE
資料タイプ
学術雑誌論文
外部リンク
ISSN
17451361
掲載誌名
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS
E105-D
2
開始ページ
415
終了ページ
426
発行日
2022-02
出版者DOI
著者版フラグ
出版社版
著作権表記
(C) 2022 The Institute of Electronics, Information and Communication Engineers
掲載情報
Nishiyama Masashi, Inoue Michiko, Iwai Yoshio, et al. Gender Recognition Using a Gaze-Guided Self-Attention Mechanism Robust Against Background Bias in Training Samples. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS. 2022. E105D(2). 415-426. doi:10.1587/transinf.2021edp7117
部局名
工学部・工学研究科
言語
英語
Web of Science Key ut
WOS:000748957000024