[Defense] Action Labeling in Images and Videos
Monday, December 13, 2021
8:00 am - 10:00 am
will defend his proposal
Action Labeling in Images and Videos
Deep learning models that attempt to categorize visual content can benefit from being trained with additional information that may be, or may not be, available during deployment. To this end, this dissertation aims to propose methods inspired by the “Learning Using Privileged Information” framework and multimodal data fusion for improving the performance of deep learning models. The proposed methods are assessed for the problems of: (i) recognizing carrying actions in “visible spectrum” and “near-infrared” images, as well as (ii) rating the age appropriateness and detecting comic mischief in online videos. The experimental results demonstrated the effectiveness of the proposed methods, in four new datasets that were introduced, within the context of this work to address the challenges of the aforementioned problems.
8:00 AM - 10:00 AM CT
Online via MSFT Teams (link TBA)
Dr. Ioannis Kakadiaris, dissertation advisor
Faculty, students and the general public are invited.