Adaptive and constrained algorithms for inverse compositional active appearance model fitting

Papandreou, G; Maragos, P

dc.contributor.author	Papandreou, G	en
dc.contributor.author	Maragos, P	en
dc.date.accessioned	2014-03-01T02:45:05Z
dc.date.available	2014-03-01T02:45:05Z
dc.date.issued	2008	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32139
dc.subject	Active Appearance Model	en
dc.subject	High Efficiency	en
dc.subject	Image Alignment	en
dc.subject	Image Synthesis	en
dc.subject	Numerical Algorithm	en
dc.subject	Parametric Model	en
dc.subject	Prior Information	en
dc.subject.other	Algorithms	en
dc.subject.other	Artificial intelligence	en
dc.subject.other	Computer vision	en
dc.subject.other	Deformation	en
dc.subject.other	Face recognition	en
dc.subject.other	Feature extraction	en
dc.subject.other	Image processing	en
dc.subject.other	Pattern recognition	en
dc.subject.other	Real time systems	en
dc.subject.other	Active Appearance Model	en
dc.subject.other	Active appearance models	en
dc.subject.other	Analysis problems	en
dc.subject.other	Constrained algorithms	en
dc.subject.other	Data-sets	en
dc.subject.other	Efficient implementation	en
dc.subject.other	Fitting algorithms	en
dc.subject.other	High-efficiency	en
dc.subject.other	Human faces	en
dc.subject.other	Image alignment	en
dc.subject.other	Image synthesis	en
dc.subject.other	Matching algorithms	en
dc.subject.other	Object appearance	en
dc.subject.other	Parametric modeling	en
dc.subject.other	Prior information	en
dc.subject.other	Adaptive algorithms	en
dc.title	Adaptive and constrained algorithms for inverse compositional active appearance model fitting	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/CVPR.2008.4587540	en
heal.identifier.secondary	http://dx.doi.org/10.1109/CVPR.2008.4587540	en
heal.identifier.secondary	4587540	en
heal.publicationDate	2008	en
heal.abstract	Parametric models of shape and texture such as Active Appearance Models (AAMs) are diverse tools for deformable object appearance modeling and have found important applications in both image synthesis and analysis problems. Among the numerous algorithms that have been proposed for AAM fitting, those based on the inverse-compositional image alignment technique have recently received considerable attention due to their potential for high efficiency. However, existing fitting algorithms perform poorly when used in conjunction with models exhibiting significant appearance variation, such as AAMs trained on multiple-subject human face images. We introduce two enhancements to inverse-compositional AAM matching algorithms in order to overcome this limitation. First, we propose fitting algorithm adaptation, by means of (a) fitting matrix adjustment and (b) AAM mean template update. Second, we show how prior information can be incorporated and constrain the AAM fitting process. The inverse-compositional nature of the algorithm allows efficient implementation of these enhancements. Both techniques substantially improve AAM fitting performance, as demonstrated with experiments on publicly available multi-person face datasets. ©2008 IEEE.	en
heal.journalName	26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR	en
dc.identifier.doi	10.1109/CVPR.2008.4587540	en