next up previous
Next: Tästä dokumentista ...

3D Model-based Object Recognition and Segmentation in Cluttered Scenes

Guest Lecture by Ass.Prof. Mohammed Bennamoun
School of Computer Science and Software Engineering
The University of Western Australia, Australia

Friday May 19, 2006, at 11.00 a.m.
TKK Institute of Photogrammetry and Remote Sensing
Lecture Hall M1, TKK Main Building, Otakaari 1, Espoo


Viewpoint independent recognition of free-form objects and their segmentation in the presence of clutter and occlusions is a challenging task. I will present a novel 3D model-based algorithm which performs this task automatically and efficiently. I will first focus on our automatic 3D model database construction algorithm. The 3D model of a free-form object is constructed during an offline phase from its multiple unordered range images (views) acquired from different viewpoints. These views are then converted into our tensor representation. Tensors are local shape descriptors which are matched to find correspondences between the views. Our multiview correspondence algorithm automatically establishes correspondences between the unordered views by simultaneously matching the tensors of a view with the tensors of the remaining views using a hash table based voting scheme. The result is a spanning tree graph of relative transformations between the views. The graph is used to register all the views which are then integrated and reconstructed to form a seamless 3D model. These models along with their tensor representations constitute the model library which is used for model-based object recognition and segmentation in the presence of clutter and occlusions as detailed in the second part of the talk. We demonstrate the robustness of our automatic multiview correspondence algorithm with respect to a number of important criteria including resolution, required number of tensors, efficiency and noise.

Short Biography:

Mohammed Bennamoun received his M.Sc. from Queen's University, Kingston, Canada in the area of Control Theory, and his PhD from Queen's /Q.U.T in Brisbane, Australia in the area of Computer Vision. He lectured Robotics at Queen's, and then joined QUT in 1993 as an Associate Lecturer. He then became a Lecturer in 1996 and a Senior Lecturer in 1998 at QUT. In Jan. 2003, he joined the Department of Computer Science and Software Engineering at The University of Western Australia as an Associate Professor. He was also the Director of a research Centre from 1998-2002. He is the co-author of the book ``Object Recognition: Fundamentals and Case Studies'', Springer-Verlag, 2001. He published over 90 journal and conference publications. He served as a guest editor for a couple of special issues in International journals, such as the International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI). He was selected to give conference tutorials at the European Conference on Computer Vision (ECCV'2002) and the International Conference on Acoustics Speech and Signal Processing (IEEE ICASSP) in 2003. He organized several special sessions for conferences; the latest was for the IEEE International Conference in Image Processing (IEEE ICIP) held in Singapore in 2004. He was on the program committee of many conferences including 3D Digital Imaging and Modeling (3DIM 2005). He also contributed in the organisation of many local and international conferences. His areas of interest include control theory, robotics, obstacle avoidance, object recognition, artificial neural networks, signal/image processing and computer vision. He served as a visiting Professor at the University of Edinburgh, The University of Paris 13, The University of Bourgogne and he is currently a visiting Professor at the Helsinki University of Technology.

next up previous
Next: Tästä dokumentista ...