SEMINAR: Polar Codes with Higher-Order Memory, by Dr. Hüseyin Afşer (June 16, 2017)
July 18, 2017
Dr. Çağlar Aytekin will give a seminar on Wednesday, July 19, 2017 at 13:30 in the Conference Hall of our department. The title of the talk is 'Quantum Mechanics and Spectral Graph Theory'.
All interested are invited.
The abstract and a short biography of the speaker are as follows.
The focus of the talk will be on Dr. Aytekin's recent studies on establishing a link between Quantum Mechanics and Spectral Graph Theory which have led to a state-of-the-art Computer Vision method. A tentative outline of the talk is as follows. First, an investigation of a direct application of Schrodinger's Equation to Computer Vision will be analyzed and it will be shown that this analysis leads to a link between the principals of Quantum Mechanics and Spectral Graph Theory. Next, it will be shown how this link is exploited to come up with Quantum Cuts: a spectral graph partitioning method for foreground detection from images and videos. Finally, extensions and improvements on Quantum Cuts will be explained and the applications of the method on salient object detection and object proposals will be discussed.
Dr. Caglar Aytekin received his Bachelor of Science Degree from Electrical and Electronics Engineering Department, Middle East Technical University in 2008. He has obtained his Master of Science degree from the same department with Signal Processing major in 2011. He received the Ph.D. degree from Signal Processing Department, Tampere University of Technology in 2016. He has been awarded with the IBM Best Student Paper Award in International Conference on Pattern Recognition-2014 for his work entitled: 'Automatic Object Segmentation by Quantum Cuts'. He has completed his Ph.D. with distinction award given by Tampere University of Technology. He has also received the Best Ph.D Thesis Award for 2015-2016 years in all Nordic countries (Finland, Sweden, Norway and Denmark) in the fields of pattern recognition and image analysis, given by Scandinavian Conference on Image Analysis Committee. His research interests cover computer vision, pattern recognition, machine learning, graph theory and application of quantum mechanics to these fields. He has authored more than 20 published/accepted articles in related areas. During his Ph.D., he has visited the Institute of Computer Graphics and Vision, Technical University of Graz and Computer Vision Lab in Hacettepe University.