Computer Science Colloquium

Reconstruction of a complex three dimensional rigid object from its two dimensional images is a challenging computer vision problem under general imaging conditions. Without a priori information about the imaging environment, it becomes very difficult to infer the 3D structure of the captured rigid object. For practical 3D reconstruction solutions, the problem can be simplified by using controlled imaging environments. Silhouette based reconstruction is a simple and robust technique for 3D volume estimation of an object. However, it has drawbacks. We use multi-baseline stereo information along with the photoconsistency principle to overcome the problems related to this technique. The algorithm basically carves the excess volume elements using the multi-baseline stereo information. In an alternative solution, we represent the surface of the model as a triangular mesh and the model is deformed such that an energy function is minimized.

Dr. Volkan Atalay is a faculty member at the Department of Computer Engineering, Middle East Technical University, Ankara, Turkey. Previously, he was a visiting scholar at the New Jersey Institute of Technology. His research interests include computer vision, document analysis and neural networks. He received a BSc and an MSc in electrical engineering from Middle East Technical University and a PhD in computer science from the Université René Descartes-Paris V, Paris, France. He is a member of the IEEE Computer Society and the Turkish Pattern Recognition and Image.