3D Human posture estimation and tracking by using vectorization and time series prediction = 시연속 예측과 벡터화 방법을 이용한 3차원 사람의 동작 추정 및 추적

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 646
  • Download : 0
The tracking and pose estimation of people in video made challenging problem due to the variability of human appearance. In this thesis, we research on articulated 3D human motion and pose estimation. However the human posture recognition has many dicult problems to be solved, because the human body model is deformable object, besides we have to process the complexity of natural scenes and the high dimensionality of articulated body models. Also the several problems such as texture of clothes, lighting change, illumination, various colors made the problem harder. Recent approaches to estimating and tracking human posture exploit articulated human body models in which the body is viewed as a kinematic tree. Since the particle filter has been proven as very successful algorithm for non-linear and non-gaussian estimation problem, the particle filtering was alternative method for human body tracking. Our goal is to develop a modified particle filter which is shown to be effective at searching the high-dimensional configuration spaces (40 dimensions) by using time series prediction. Our results was suggested in constrained laboratory environments and showed our methods perform quite well than others. Multiple cameras and background subtraction, however, are required to achieve reliable tracking performance. In this thesis, we proposed the admix vector and time series vector (TSV) particle filter. The proposed likelihood which is admix vector can reduce the computation time and solve the asymmetric problem at the same time. To solve the problem of the robust tracking of fast moving object, the time series prediction was employed to predict states following regression information of RBF neural network.
Advisors
Lee, Ju-Jangresearcher이주장
Description
한국과학기술원 : 로봇공학학제전공,
Publisher
한국과학기술원
Issue Date
2011
Identifier
482719/325007  / 020094219
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 로봇공학학제전공, 2011.8, [ vi, 53 p. ]

Keywords

Particle filter; Time series prediction; Kinematic tree; Multiple camera; 극소 필터; 시연속 예측; 키네마틱 트리; 다중 카메라; RBF 뉴럴넷; RBF neural network

URI
http://hdl.handle.net/10203/182408
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=482719&flag=dissertation
Appears in Collection
RE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0