2026-04-12T20:32:43Z https://ipsj.ixsq.nii.ac.jp/oai

oai:ipsj.ixsq.nii.ac.jp:00209846 2025-01-19T18:22:18Z 1164:4619:10416:10532

クロスビュー・非局所ニューラルネットワークによる自己視点映像と固定視点映像間の共通特徴量の学習 Cross-view Non-local Neural Networks for Joint Representation Learning between First and Third Person Videos シュ, テツコウ 菅野, 裕介 佐藤, 洋一 Zhehao, Zhu Yusuke, Sugano Yoichi, Sato セッション6-2 This paper introduces a cross-view non-local neural network to learn joint representations for understanding human activities from videos captured by wearable and fixed cameras. The key element is a non-local model to extract and enhance the global visual feature similarity across the views while reducing dissimilarity. The proposed method achieves a state-of-the-art performance on a cross-view action recognition benchmark dataset. technical report 情報処理学会 2021-02-25 application/pdf 研究報告コンピュータビジョンとイメージメディア（CVIM） 48 2021-CVIM-225 1 6 2188-8701 AA11131797 https://ipsj.ixsq.nii.ac.jp/record/209846/files/IPSJ-CVIM21225048.pdf eng