Abstract | ||
---|---|---|
This paper proposes a new method for achieving precise video segmentation without any supervision or interaction. The main contributions of this report include 1) the introduction of fully automatic segmentation based on the maximum a posteriori (MAP) estimation of the Markov random field (MRF) with graph cuts and saliency-driven priors and 2) the updating of priors and feature likelihoods by integrating the previous segmentation results and the currently estimated saliency-based visual attention. Test results indicate that our new method precisely extracts probable regions from videos without any supervised interactions. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICME.2009.5202577 | ICME |
Keywords | Field | DocType |
main contribution,markov random field,saliency-based video segmentation,extracts probable region,automatic segmentation,sequentially updated prior,previous segmentation result,precise video segmentation,new method,graph cut,saliency-driven prior,feature likelihood,random processes,image segmentation,maximum a posteriori estimation,saliency,kalman filter,graph cuts,visualization,kalman filters,graph theory,maximum likelihood estimation,markov processes,pixel | Cut,Graph theory,Computer vision,Scale-space segmentation,Pattern recognition,Computer science,Salience (neuroscience),Segmentation,Markov random field,Image segmentation,Artificial intelligence,Maximum a posteriori estimation | Conference |
ISSN | Citations | PageRank |
1945-7871 | 53 | 1.43 |
References | Authors | |
12 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ken Fukuchi | 1 | 75 | 2.82 |
Kouji Miyazato | 2 | 65 | 2.58 |
Akisato Kimura | 3 | 244 | 28.03 |
Shigeru Takagi | 4 | 53 | 1.43 |
Junji Yamato | 5 | 1120 | 165.72 |