CS Other Presentations

Department of Computer Science - University of Cyprus

Besides Colloquiums, the Department of Computer Science at the University of Cyprus also holds Other Presentations (Research Seminars, PhD Defenses, Short Term Courses, Demonstrations, etc.). These presentations are given by scientists who aim to present preliminary results of their research work and/or other technical material. Other Presentations serve as a forum for educating Computer Science students and related announcements are disseminated to the Department of Computer Science (i.e., the csall list):
rss RSS Directions Directions

Presentations Coordinator: Demetris Zeinalipour

PhD Defense: Physically-Based Probabilistic Image Segmentation, Mr. Nikolas Ladas (University of Cyprus, Cyprus), Tuesday, January 28, 2020, 09.30-10.30 EET.

The Department of Computer Science at the University of Cyprus cordially invites you to the PhD Defense entitled:

Physically-Based Probabilistic Image Segmentation

Speaker: Mr. Nikolas Ladas
Affiliation: University of Cyprus, Cyprus
Category: PhD Defense
Location: Room 148, Faculty of Pure and Applied Sciences (FST-01), 1 University Avenue, 2109 Nicosia, Cyprus (directions)
Date: Tuesday, January 28, 2020
Time: 09.30-10.30 EET
Host: Prof. Constantinos Pattichis (pattichi-AT-cs.ucy.ac.cy)
URL: https://www.cs.ucy.ac.cy/colloquium/presentations.php?speaker=cs.ucy.pres.2020.ladas

Image segmentation is a vital component of many Computer Vision algorithms including object identification, tracking, and image manipulation. These algorithms power various high impact applications such as image editing, composition, film post-processing, autonomous driving and virtual/augmented reality. Although image segmentation is a mature field, existing algorithms often fail when the input -images or video- contains strong illumination effects such as shadows and colored lighting. This thesis addresses some of these limitations by incorporating knowledge from the Computer Graphics field where complex illumination effects are well-studied. Our contribution is twofold: Firstly, we introduce a data acquisition process that utilizes high dynamic range imaging to capture the illumination of the scene. Based on this process we have developed an illumination normalization algorithm that improves tracking performance in cases where the scene illumination changes rapidly. The second contribution of this thesis is two algorithms for the segmentation of images into background and foreground regions. The proposed algorithms utilize a physically-based formulation of scene appearance which explicitly models the formation of shadows originating from multiple, possibly colored, light sources. This formulation enables a probabilistic model to distinguish between shadows and foreground objects in challenging images, such as those lit by colored lights. The proposed methods are efficient, general, and robust.

Short Bio:
Nikolas Ladas is a Ph.D. candidate at the Computer Science Department under the supervision of professor Yiorgos Chrysanthou. His research interests lie in the areas of Computer Graphics and Computer Vision and specifically in inverse illumination and scene segmentation. Nikolas is also the co-founder of Ten Ton Train LTD which develops technology for commercial and serious games. He received his BSc and MSc from the University of Cyprus.

  Other Presentations Web: https://www.cs.ucy.ac.cy/colloquium/presentations.php
  Colloquia Web: https://www.cs.ucy.ac.cy/colloquium/
  Calendar: https://www.cs.ucy.ac.cy/colloquium/schedule/cs.ucy.pres.2020.ladas.ics