Title :
Tracking non-rigid objects in complex scenes
Author :
Huttenlocher, Daniel P. ; Noh, Jae J. ; Rucklidge, William J.
Author_Institution :
Dept. of Comput. Sci., Cornell Univ., Ithaca, NY, USA
Abstract :
The authors describe a model-based method for tracking nonrigid objects moving in a complex scene. The method operates by extracting two-dimensional models of an object from a sequence of images. The basic idea underlying the technique is to decompose the image of a solid object moving in space into two components: a two-dimensional motion and a two-dimensional shape change. The motion component is factored out and the shape change is represented explicitly by a sequence of two-dimensional models, one corresponding to each image frame. The major assumption underlying the method is that the two-dimensional shape of an object will change slowly from one frame to the next. There is no assumption, however, that the two-dimensional image motion between successive frames will be small
Keywords :
image recognition; motion estimation; object detection; complex scenes; model-based method; nonrigid objects tracking; solid object; two-dimensional models; two-dimensional motion; two-dimensional shape change; Computer science; Computer vision; Contracts; Data mining; Image sequences; Layout; Pixel; Shape; Solids; Tracking;
Conference_Titel :
Computer Vision, 1993. Proceedings., Fourth International Conference on
Conference_Location :
Berlin
Print_ISBN :
0-8186-3870-2
DOI :
10.1109/ICCV.1993.378231