A pseudo node based graph convolution network for emotion perception from gait

• Item type: SIG Technical Reports(1)
• 公開日: 2022-06-20
• タイトル: A pseudo node based graph convolution network for emotion perception from gait
• タイトル / 言語: en
• タイトル: A pseudo node based graph convolution network for emotion perception from gait
• 言語: eng
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Ritsumeikan University
• 著者所属: Ritsumeikan University
• 著者所属: Fujita Health University
• 著者所属: Zhejiang university
• 著者所属: Ritsumeikan University
• 著者所属(英): en
• 著者所属(英): Ritsumeikan University
• 著者所属(英): en
• 著者所属(英): Ritsumeikan University
• 著者所属(英): en
• 著者所属(英): Fujita Health University
• 著者所属(英): en
• 著者所属(英): Zhejiang university
• 著者所属(英): en
• 著者所属(英): Ritsumeikan University
• 著者名: Shurong, Chai; Jiaqing, Liu; Tomoko, Tateyama; Lanfen, Lin; Yen-Wei, Chen
• 著者名(英): Shurong, Chai; Jiaqing, Liu; Tomoko, Tateyama; Lanfen, Lin; Yen-Wei, Chen
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Recently, emotion recognition task has attracted extensive attention in the field of human-computer interaction. With the advancement of Graph Convolutional Networks (GCNs), human gait can be effectively recognized in more complex backgrounds. According to the anatomy of the human body, central torso joints play a key role in GCNs-based human gait recognition systems, instead of the body’s marginal limb joints. As a result, there is a major issue of receptive field imbalance. In this study, we propose a method for perceiving emotions based on the human gait skeleton. We present a novel pseudo node strategy that connects all natural human body joints to alleviate the receptive field imbalance problem. The result of the experiment show that our proposed method outperforms than existing skeleton-based methods.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Recently, emotion recognition task has attracted extensive attention in the field of human-computer interaction. With the advancement of Graph Convolutional Networks (GCNs), human gait can be effectively recognized in more complex backgrounds. According to the anatomy of the human body, central torso joints play a key role in GCNs-based human gait recognition systems, instead of the body’s marginal limb joints. As a result, there is a major issue of receptive field imbalance. In this study, we propose a method for perceiving emotions based on the human gait skeleton. We present a novel pseudo node strategy that connects all natural human body joints to alleviate the receptive field imbalance problem. The result of the experiment show that our proposed method outperforms than existing skeleton-based methods.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10505667
• 書誌情報: 研究報告数理モデル化と問題解決（MPS）; 巻 2022-MPS-138, 号 19, p. 1-3, 発行日 2022-06-20
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8833
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会