Flexible top-view human pose estimation for detection system via CNN

Ryuji Go, Yoshimitsu Aoki

    研究成果: Conference contribution

    抄録

    We propose the DeepPose-based pose estimation system that is flexible with the change of bounding-box range for top-view images. Our purpose is to link person detection system and pose estimation system. We introduce Bounding-box Curriculum Learning (BCL) and Recurrent Pose Estimation (RPE). BCL is a learning technique of CNN inspired from Curriculum Learning. RPE is a recurrent process of pose estimation that fixes the bounding-box range in response to the estimated results. We show the effect of proposed methods compared to normal learned CNN-based pose estimator on our original top-view dataset.

    本文言語English
    ホスト出版物のタイトル2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016
    出版社Institute of Electrical and Electronics Engineers Inc.
    ISBN(電子版)9781509023332
    DOI
    出版ステータスPublished - 2016 12 27
    イベント5th IEEE Global Conference on Consumer Electronics, GCCE 2016 - Kyoto, Japan
    継続期間: 2016 10 112016 10 14

    Other

    Other5th IEEE Global Conference on Consumer Electronics, GCCE 2016
    国/地域Japan
    CityKyoto
    Period16/10/1116/10/14

    ASJC Scopus subject areas

    • 信号処理
    • 電子工学および電気工学
    • コンピュータ サイエンスの応用
    • ハードウェアとアーキテクチャ
    • 器械工学

    フィンガープリント

    「Flexible top-view human pose estimation for detection system via CNN」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

    引用スタイル