DocumentCode :
3405007
Title :
Optical flow estimation with adaptive convolution kernel prior on discrete framework
Author :
Lee, Kyong Joon ; Kwon, Dongjin ; Yun, Dong, II ; Lee, Sang Uk
Author_Institution :
Sch. of EECS, Seoul Nat´´l Univ., Seoul, South Korea
fYear :
2010
fDate :
13-18 June 2010
Firstpage :
2504
Lastpage :
2511
Abstract :
We present a new energy model for optical flow estimation on discrete MRF framework. The proposed model yields discrete analog to the prevailing model with diffusion tensor-based regularizer, which has been optimized by variational approach. Inspired from the fact that the regularization process works as a convolution kernel filtering, we formulate the difference between original flow and filtered flow as a smoothness prior. Then the discrete framework enables us to employ a robust penalizer less concerning convexity and differentiability of the energy function. In addition, we provide a new kernel design based on the bilateral filter, adaptively controlling intensity variance according to the local statistics. The proposed kernel simultaneously addresses over-segmentation and over-smoothing problems, which is hard to achieve by tuning parameters. Involving a complex graph structure with large label sets, this work also presents a strategy to efficiently reduce memory requirement and computational time to a tolerable state. Experimental result shows the proposed method yields plausible results on the various data sets including large displacement and textured region.
Keywords :
filtering theory; image sequences; adaptive convolution kernel; bilateral filter; complex graph structure; convolution kernel filtering; diffusion tensor-based regularizer; discrete MRF framework; energy function; intensity variance; local statistics; optical flow estimation; over-segmentation problems; over-smoothing problems; Adaptive optics; Brightness; Convolution; Costs; Filtering; Image motion analysis; Kernel; Optical filters; Robustness; Statistics;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on
Conference_Location :
San Francisco, CA
ISSN :
1063-6919
Print_ISBN :
978-1-4244-6984-0
Type :
conf
DOI :
10.1109/CVPR.2010.5539953
Filename :
5539953
Link To Document :
بازگشت