DocumentCode :
999899
Title :
Figure-ground segmentation from occlusion
Author :
Aguiar, Pedro M. Q. ; Moura, José M F
Author_Institution :
Inst. for Syst. & Robotics, Inst. Superior Tecnico, Lisbon, Portugal
Volume :
14
Issue :
8
fYear :
2005
Firstpage :
1109
Lastpage :
1124
Abstract :
Layered video representations are increasingly popular; see for a recent review. Segmentation of moving objects is a key step for automating such representations. Current motion segmentation methods either fail to segment moving objects in low-textured regions or are computationally very expensive. This paper presents a computationally simple algorithm that segments moving objects, even in low-texture/low-contrast scenes. Our method infers the moving object templates directly from the image intensity values, rather than computing the motion field as an intermediate step. Our model takes into account the rigidity of the moving object and the occlusion of the background by the moving object. We formulate the segmentation problem as the minimization of a penalized likelihood cost function and present an algorithm to estimate all the unknown parameters: the motions, the template of the moving object, and the intensity levels of the object and of the background pixels. The cost function combines a maximum likelihood estimation term with a term that penalizes large templates. The minimization algorithm performs two alternate steps for which we derive closed-form solutions. Relaxation improves the convergence even when low texture makes it very challenging to segment the moving object from the background. Experiments demonstrate the good performance of our method.
Keywords :
image representation; image segmentation; maximum likelihood estimation; minimisation; video signal processing; background occlusion; figure-ground segmentation; image intensity values; layered video representations; low-texture-low-contrast scenes; maximum likelihood estimation; minimization algorithm; motion segmentation methods; moving object rigidity; moving object segmentation; penalized likelihood cost function; relaxation; Closed-form solution; Computer vision; Cost function; Image segmentation; Layout; Maximum likelihood estimation; Minimization methods; Motion estimation; Motion segmentation; Video sequences; Layered video representations; motion; occlusion; penalized likelihood; rigidity; segmentation; Algorithms; Computer Simulation; Image Enhancement; Image Interpretation, Computer-Assisted; Information Storage and Retrieval; Models, Statistical; Movement; Numerical Analysis, Computer-Assisted; Pattern Recognition, Automated; Signal Processing, Computer-Assisted; Subtraction Technique; Video Recording;
fLanguage :
English
Journal_Title :
Image Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1057-7149
Type :
jour
DOI :
10.1109/TIP.2005.851712
Filename :
1468196
Link To Document :
بازگشت