Tracking complex objects using graphical object models
作者： Leonid Sigal , Ying Zhu , Dorin Comaniciu , Michael Black
DOI: 10.1007/978-3-540-69866-1_17
关键词: Video tracking 、 Viola–Jones object detection framework 、 Pattern recognition 、 Object-oriented design 、 Artificial intelligence 、 Computer science 、 Object detection 、 Object-class detection 、 Graphical model 、 Object model 、 Method 、 Computer vision
摘要: We present a probabilistic framework for component-based automatic detection and tracking of objects in video. represent as spatio-temporal two-layer graphical models, where each node corresponds to an object or component at given time, the edges correspond learned spatial temporal constraints. Object is formulated inference over directed loopy graph, solved with non-parametric belief propagation. This type model allows object-detection make use consistency (over arbitrarily sized window), facilitates robust object. The two layer structure entire well individual components. AdaBoost detectors are used define likelihood form proposal distributions Proposal provide 'bottom-up' information that incorporated into process, enabling tracking. illustrate our method by detecting classes objects, vehicles pedestrians, video sequences collected using single grayscale uncalibrated car-mounted moving camera.
brown.edu
springer.com
mpg.de 
sci-hub.st 参考文章(21)
Arnaud Doucet, Nando Freitas, Neil Gordon, An Introduction to Sequential Monte Carlo Methods Sequential Monte Carlo Methods in Practice. pp. 3- 14 ,(2001) , 10.1007/978-1-4757-3437-9_1
Sequential Monte Carlo methods in practice Springer. ,(2001) , 10.1007/978-1-4757-3437-9
Michael J. Black, Yaser Yacoob, Recognizing Facial Expressions in Image Sequences Using Local Parameterized Models of Image Motion International Journal of Computer Vision. ,vol. 25, pp. 23- 48 ,(1997) , 10.1023/A:1007977618277
Binglong Xie, Dorin Comaniciu, Visvanathan Ramesh, Markus Simon, Terrance Boult, Component Fusion for Face Detection in the Presence of Heteroscedastic Noise joint pattern recognition symposium. pp. 434- 441 ,(2003) , 10.1007/978-3-540-45243-0_56
James M. Coughlan, Sabino J. Ferreira, Finding Deformable Shapes Using Loopy Belief Propagation european conference on computer vision. pp. 453- 468 ,(2002) , 10.1007/3-540-47977-5_30
Michael C. Burl, Markus Weber, Pietro Perona, A Probabilistic Approach to Object Recognition Using Local Photometry and Global Geometry european conference on computer vision. pp. 628- 641 ,(1998) , 10.1007/BFB0054769
Paul Viola, Michael J. Jones, Daniel Snow, Detecting Pedestrians Using Patterns of Motion and Appearance international conference on computer vision. ,vol. 63, pp. 734- 741 ,(2003) , 10.1007/S11263-005-6644-8
Michael Irwin Jordan, Terrence Joseph Sejnowski, None, Graphical Models: Foundations of Neural Computation Pattern Analysis and Applications. ,vol. 5, pp. 401- 402 ,(2001) , 10.1007/S100440200036
P.F. Felzenszwalb, D.P. Huttenlocher, Efficient matching of pictorial structures computer vision and pattern recognition. ,vol. 2, pp. 2066- 2066 ,(2000) , 10.1109/CVPR.2000.854739
A Torralba, None, Using the Forest to See the Trees: A Graphical Model Relating Features, Objects, and Scenes neural information processing systems. ,vol. 16, pp. 1499- 1506 ,(2003)