|
Download: |
Sparse Representations, Dictionaries and Domain Adaptation Methods for Image and Video-based Face RecognitionProf. Rama Chellappa Feature extraction or representation of patterns and adaptation of classifiers designed using training data to be effective on testing data are two fundamental problems in image and video-based face recognition. In this talk, I will discuss new solutions to these problems based on theories of sparse representations, dictionary learning and domain adaptation with applications in image and video-based face recognition. Specifically, I will discuss methods for representing images and videos using linear and non-linear dictionaries. I will then discuss methods for adapting the dictionaries for addressing shifts in data distributions due to changes in pose, illuminations, spatio-temporal sampling and blur with applications to recognition of faces. |
 |
