本資料は、掲載誌(URI)等のリンク先にある学位授与機関のWebサイトやCiNii Dissertationsから、本文を自由に閲覧できる場合があります。
博士論文
Study on Similarity Measure of Human Body Motion
国立国会図書館館内限定公開
収録元データベースで確認する
国立国会図書館デジタルコレクション
デジタルデータあり
Study on Similarity Measure of Human Body Motion
- 国立国会図書館永続的識別子
- info:ndljp/pid/13332034
国立国会図書館での利用に関する注記
資料に関する注記
一般注記:
- type:ThesisThe study introduces non-model-based approaches for measuring the similarity of human body motions through measuring the dissimilarity of m...
書店で探す
障害者向け資料で読む
書店で探す
障害者向け資料で読む
書誌情報
この資料の詳細や典拠(同じ主題の資料を指すキーワード、著者名)等を確認できます。
デジタル
- 資料種別
- 博士論文
- 著者・編者
- PHAM TRUONG HONG NGAN
- 出版年月日等
- 2020-09-24
- 出版年(W3CDTF)
- 2020-09-24
- 並列タイトル等
- 身体動作の類似度に関する研究
- 授与機関名
- 京都工芸繊維大学
- 授与年月日
- 2020-09-24
- 授与年月日(W3CDTF)
- 2020-09-24
- 報告番号
- 甲第974号
- 学位
- 博士(工学)
- 本文の言語コード
- eng
- 対象利用者
- 一般
- 一般注記
- type:ThesisThe study introduces non-model-based approaches for measuring the similarity of human body motions through measuring the dissimilarity of marker trajectories in human body motions. The proposed methods in the research are independent of the biological characteristics of the human body. In general terms, the human body motion generates three-dimensional (3D) curves representing the direction changes of marker trajectories. The first or original method is based on the dissimilarity measure for 3D curves. In detail, the dissimilarity of human body motions is estimated by applying the dissimilarity measure of the 3D curves for the dissimilarity of 3D curves representing the trajectories of markers and is implemented in five steps. Firstly, the motion speed is adjusted. Next, human postures in motion are normalized. Then, motion paths of markers are represented by direction sequences. After that, chain codes of direction sequences are obtained. Finally, the dissimilarity of human body movements is calculated based on the obtained chain codes. The roles of markers in motion are also taken into account. In addition, two options of the spatial quantization method for the human active space in a body motion based on the Space Harmony Theory is also mentioned. In order to reduce the conducted time of the original method, the extended method to calculate the dissimilarity of human body motions through sub-motions is also proposed. In the extended method, two options of the pause derivation method for a human body motion are introduced. Human body motions are split into sub-motions by using the pause waveforms obtaining from the pause derivation method. Because the sub-motions are considered as new motions, the dissimilarity of sub-motions can be calculated by using the original method. Both the original method and the extended method are experimentally evaluated by using the statistical test. Experimental results proved that both two methods are effective in the estimation of dissimilarity of the human body motions with the appropriate Split value. The result shows that the extended method improved the conducted time and can be used in real-time applications. Furthermore, the first option of the extended method is effective in enhancing the performing time. The proposed methods are proved effective in measuring the similarity of human body motions.
- 国立国会図書館永続的識別子
- info:ndljp/pid/13332034
- コレクション(共通)
- コレクション(障害者向け資料:レベル1)
- コレクション(個別)
- 国立国会図書館デジタルコレクション > デジタル化資料 > 博士論文
- 収集根拠
- 博士論文(自動収集)
- 受理日(W3CDTF)
- 2024-02-02T16:16:38+09:00
- 記録形式(IMT)
- application/pdf
- オンライン閲覧公開範囲
- 国立国会図書館内限定公開
- デジタル化資料送信
- 図書館・個人送信対象外
- 遠隔複写可否(NDL)
- 可
- 掲載誌(URI)
- 連携機関・データベース
- 国立国会図書館 : 国立国会図書館デジタルコレクション