Title :
Using energy flow information for video segmentation
Author :
Zhang, Zhongwei ; Liu, Guizhong ; Li, Hongliang ; Li, Yongli
Author_Institution :
Sch. of Electron. & Inf., Xi´´an Jiaotong Univ., China
Abstract :
A novel concept, the image temperature, is proposed to describe the image motion in a video. The idea comes from the revelation of the Maxwell Distribution Law in the gas molecule moving theories. Furthermore, Energy Flow Equation (EFE) is developed to estimate the motion field between images. The energy flow field estimated by the EFE presents more accurate motion field than the traditional optical flow field does. When the energy flow information is applied to the video segmentation, the experimental results indicate that the new method gives better performance than the existing method.
Keywords :
image segmentation; image sequences; motion estimation; Maxwell distribution law; energy flow equation; energy flow field; energy flow information; gas molecule; image motion field estimation; image temperature; optical flow field; video segmentation; Image motion analysis; Image segmentation; Maxwell equations; Maxwell-Boltzmann distribution; Motion analysis; Motion estimation; Physics; Pixel; Temperature distribution; Video sequences;
Conference_Titel :
Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on
Conference_Location :
Nanjing
Print_ISBN :
0-7803-7702-8
DOI :
10.1109/ICNNSP.2003.1281085