In Partial Fulfillment of the Requirements for the Degree of Master of Science
will defend his thesis
Analyzing Notions of Artistic Style Using Computer Vision Techniques
In the last couple of years we have seen computer vision make significant strides in the area of artistic style transfer, and a few attempts have also been made to extract and define the style signature of various artists. But most of these endeavors have been limited by treating a creative task such as painting and critiquing style like a traditional machine learning problem. In this study, we attempts to address and overcome the limitations of such an approach.We aim to take first steps, towards building notions of artistic style similairty with applications of critiquing and valuing art in mind. We build features for artistic style in a bottom up fasion, rooted deeply in the domain knowledge of the problem we are trying to address. We assess the efficacy of each of the features in capturing their respective stylistic elements. We establish and validate the rank and weighting of individual style features/elements to develop a cumulative measure of style similarity. We present the results as a comparison to previous efforts as well as against historically known facts about the domain.
Date: Monday, April 23, 2018
Time: 1:00 PM
Place: PGH 550
Advisor: Prof. Shishir Shah
Faculty, students, and the general public are invited.