Synergy between object recognition and image segmentation using the expectation-maximization algorithm

Kokkinos, I; Maragos, P

dc.contributor.author	Kokkinos, I	en
dc.contributor.author	Maragos, P	en
dc.date.accessioned	2014-03-01T01:32:01Z
dc.date.available	2014-03-01T01:32:01Z
dc.date.issued	2009	en
dc.identifier.issn	0162-8828	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/20021
dc.subject	Active Appearance Models	en
dc.subject	Curve evolution	en
dc.subject	Enerative models	en
dc.subject	Expectation Maximization	en
dc.subject	Image segmentation	en
dc.subject	Object recognition	en
dc.subject	Top--down segmentation	en
dc.subject.classification	Computer Science, Artificial Intelligence	en
dc.subject.classification	Engineering, Electrical & Electronic	en
dc.subject.other	Active Appearance Models	en
dc.subject.other	Curve evolution	en
dc.subject.other	Enerative models	en
dc.subject.other	Expectation Maximization	en
dc.subject.other	Expectation-maximization algorithms	en
dc.subject.other	Fitting equations	en
dc.subject.other	Image segments	en
dc.subject.other	Joint segmentation	en
dc.subject.other	Object Detection	en
dc.subject.other	Object model	en
dc.subject.other	Over segmentation	en
dc.subject.other	Segmentation algorithms	en
dc.subject.other	Segmentation informations	en
dc.subject.other	Segmentation results	en
dc.subject.other	Shape priors	en
dc.subject.other	Systematic experiment	en
dc.subject.other	Topdown	en
dc.subject.other	Algorithms	en
dc.subject.other	Digital image storage	en
dc.subject.other	Feature extraction	en
dc.subject.other	Image reconstruction	en
dc.subject.other	Image segmentation	en
dc.subject.other	Maximum principle	en
dc.subject.other	Optimization	en
dc.subject.other	Object recognition	en
dc.subject.other	algorithm	en
dc.subject.other	article	en
dc.subject.other	artificial intelligence	en
dc.subject.other	automated pattern recognition	en
dc.subject.other	face	en
dc.subject.other	histology	en
dc.subject.other	human	en
dc.subject.other	image processing	en
dc.subject.other	methodology	en
dc.subject.other	roc curve	en
dc.subject.other	Algorithms	en
dc.subject.other	Artificial Intelligence	en
dc.subject.other	Face	en
dc.subject.other	Humans	en
dc.subject.other	Image Processing, Computer-Assisted	en
dc.subject.other	Pattern Recognition, Automated	en
dc.subject.other	ROC Curve	en
dc.title	Synergy between object recognition and image segmentation using the expectation-maximization algorithm	en
heal.type	journalArticle	en
heal.identifier.primary	10.1109/TPAMI.2008.158	en
heal.identifier.secondary	http://dx.doi.org/10.1109/TPAMI.2008.158	en
heal.language	English	en
heal.publicationDate	2009	en
heal.abstract	In this work, we formulate the interaction between image segmentation and object recognition in the framework of the Expectation-Maximization (EM) algorithm. We consider segmentation as the assignment of image observations to object hypotheses and phrase it as the E-step, while the M-step amounts to fitting the object models to the observations. These two tasks are performed iteratively, thereby simultaneously segmenting an image and reconstructing it in terms of objects. We model objects using Active Appearance Models (AAMs) as they capture both shape and appearance variation. During the E-step, the fidelity of the AAM predictions to the image is used to decide about assigning observations to the object. For this, we propose two top-down segmentation algorithms. The first starts with an oversegmentation of the image and then softly assigns image segments to objects, as in the common setting of EM. The second uses curve evolution to minimize a criterion derived from the variational interpretation of EM and introduces AAMs as shape priors. For the M-step, we derive AAM fitting equations that accommodate segmentation information, thereby allowing for the automated treatment of occlusions. Apart from top-down segmentation results, we provide systematic experiments on object detection that validate the merits of our joint segmentation and recognition approach. © 2009 IEEE.	en
heal.publisher	IEEE COMPUTER SOC	en
heal.journalName	IEEE Transactions on Pattern Analysis and Machine Intelligence	en
dc.identifier.doi	10.1109/TPAMI.2008.158	en
dc.identifier.isi	ISI:000267050600011	en
dc.identifier.volume	31	en
dc.identifier.issue	8	en
dc.identifier.spage	1486	en
dc.identifier.epage	1501	en