講演名 2012-03-30
Object Tracking Using Kinect and its Application in Tracking Dynamical Behaviors of Botanical Tree Leaves
,
PDFダウンロードページ PDFダウンロードページへ
抄録(和)
抄録(英) Reconstructing the dynamical behavior of a moving non-rigid object such as smoke, water and botanical tree is a tough task because it is very difficult to track the behavior of such objects due to their complicated structure and behavior. This paper studies on tracking the dynamical behaviors of a botanical tree by particle filter based methods. The proposed method uses particle filter, which is also known as Sequential Monte Carlo method, to estimate the position of the target leaf (leaf to be tracked) in each frame of the video sequences. The motion model predicts the position of the target leaf by generating different hypothesis of the target leaf's positions. The observation model corrects the prediction by evaluating each "particle" based on template matching score between the leaf model (template) and the target leaf. To increase the tracking accuracy and decrease the computation cost, a leaf is only searched for in its region of interest (ROI). The ROI position and size are updated by one frame delay method.
キーワード(和)
キーワード(英) Particle filter / Template matching / Kinect / Botanical tree / Tracking
資料番号 PRMU2011-265,HIP2011-93
発行日

研究会情報
研究会 HIP
開催期間 2012/3/22(から1日開催)
開催地(和)
開催地(英)
テーマ(和)
テーマ(英)
委員長氏名(和)
委員長氏名(英)
副委員長氏名(和)
副委員長氏名(英)
幹事氏名(和)
幹事氏名(英)
幹事補佐氏名(和)
幹事補佐氏名(英)

講演論文情報詳細
申込み研究会 Human Information Processing (HIP)
本文の言語 ENG
タイトル(和)
サブタイトル(和)
タイトル(英) Object Tracking Using Kinect and its Application in Tracking Dynamical Behaviors of Botanical Tree Leaves
サブタイトル(和)
キーワード(1)(和/英) / Particle filter
第 1 著者 氏名(和/英) / Khanh N. DO
第 1 著者 所属(和/英)
Graduate school of Global Information and Telecommunication Studies, Waseda University
発表年月日 2012-03-30
資料番号 PRMU2011-265,HIP2011-93
巻番号(vol) vol.111
号番号(no) 500
ページ範囲 pp.-
ページ数 6
発行日