Bilkent University
Department of Computer Engineering
MS Thesis Presentation


Activity Analysis for Assistive Systems


Ahmet İşcen
MS Student
Computer Engineering Department
Bilkent University

Although understanding and analyzing human actions is a popular research topic in computer vision, most of the research is focused on recognizing "ordinary" actions, such as walking and jumping. Extending these methods for more specific domains, such as assistive technologies, is not a trivial task. In most cases, these applications contain more fine-grained activities with low inter-class variance and high intra-class variance. In this thesis, we propose to use motion information from snippets, or small video intervals, in order to recognize actions from daily activities. We also show that modeling the sequential information of actions where the visual information is not distinguishable enough, improves the performance of the system. We conduct experiments in medical device usage and cooking activities domains to test the performance of our system.


DATE: 04 August, 2014, Monday @ 11:00