Seminar by Tanaya Guha
Classification and Similarity Measurement by Learning Sparse Representations
Tanaya Guha
Univ. of British Columbia, Canada
Date: Thursday, October 17th, 2013
Time: 5:00 PM
Venue: CS102.Abstract:
Sparse representation of signals has recently emerged as a major research area. It is well-known that many natural signals can be sparsely represented using a properly chosen dictionary (e.g. formed of wavelets bases). A dictionary could be complete or overcomplete depending on whether the number of bases it contains is the same or greater than the dimensionality of the given signal. Traditionally, the use of predefined dictionaries has been prevalent in sparse analysis. However, a more generalized approach is to learn the dictionary from the signal itself. Learnt dictionaries are known to outperform predefined dictionaries in several applications. We explore the application of sparse representations of signals obtained by learning overcomplete dictionaries for three applications: 1) classification of images and videos, 2) measurement of similarity between two images, and 3) assessment of perceptual quality of an image.
About the speaker:
Tanaya Guha has received her PhD (Electrical and Computer Engg) in 2013 from the University of British Columbia, Vancouver, Canada. She is broadly interested in pattern recognition and signal/image processing. Her PhD work involves learning sparse representations for classification and image similarity measurement. She was a recipient of Mensa Canada Woodhams memorial scholarship, Google Anita Borg memorial scholarship and Amazon Grace Hopper celebration scholarship. She holds an MASc degree in Electrical Engg (University of Windsor, Canada) and a BE degree in Electronics and Telecomm Engg (Bengal Engineering and Science University, Shibpore, India).