DocumentCode :
3170385
Title :
A multiresolution approach to integrated optical flow computation
Author :
Kumar, K. Sunil ; Desai, U.B.
Author_Institution :
Indian Inst. of Technol., Bombay, India
fYear :
1995
fDate :
4-6 Jul 1995
Firstpage :
852
Lastpage :
857
Abstract :
Uses simulated annealing to minimize energy functions at all resolutions. One can enhance the speed of computation by using simulated annealing at the coarsest resolution and then using the less computationally complex deterministic algorithms like graduated non-convexity, iterative conditional modes at finer resolutions, the idea being that the simulated annealing algorithm would have given a good estimate for the deterministic algorithms to perform better at finer resolution
Keywords :
computational complexity; feature extraction; image matching; image resolution; image sequences; interpolation; coarse resolution; computation time; computational complexity; deterministic algorithms; energy functions minimization; fine resolutions; graduated non-convexity; graduated nonconvexity; integrated optical flow computation; iterative conditional mode; iterative conditional modes; multiresolution approach; simulated annealing;
fLanguage :
English
Publisher :
iet
Conference_Titel :
Image Processing and its Applications, 1995., Fifth International Conference on
Conference_Location :
Edinburgh
Print_ISBN :
0-85296-642-3
Type :
conf
DOI :
10.1049/cp:19950781
Filename :
465648
Link To Document :
بازگشت