Quadtree-based eigendecomposition for pose estimation in the presence of occlusion and background clutter

Authors:
Chu-Yin Chang;Anthony A. Maciejewski;Venkataramanan Balakrishnan;Rodney G. Roberts;Kishor Saitwal
Affiliations:
Energid Technologies, 124 Mount Auburn Street, Suite 200, 02138, North Cambridge, MA, USA;Colorado State University, Department of Electrical and Computer Engineering, 124 Mount Auburn Street, Suite 200, 80523-1373, Fort Collins, CO, USA;Purdue University, Department of Electrical and Computer Engineering, 124 Mount Auburn Street, Suite 200, 47907-1285, West Lafayette, IN, USA;Florida A&M——Florida State University, Department of Electrical and Computer Engineering, 124 Mount Auburn Street, Suite 200, 32310-6046, Tallahassee, FL, USA;Colorado State University, Department of Electrical and Computer Engineering, 124 Mount Auburn Street, Suite 200, 80523-1373, Fort Collins, CO, USA
Venue:
Pattern Analysis & Applications
Year:
2007

Citing 0
Cited 2

Representing images of a rotating object with cyclic permutation for view-based pose estimation

Computer Vision and Image Understanding
Designing eigenspace manifolds: with application to object identification and pose estimation

SMC'09 Proceedings of the 2009 IEEE international conference on Systems, Man and Cybernetics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Eigendecomposition-based techniques are popular for a number of computer vision problems, e.g., object and pose estimation, because they are purely appearance based and they require few on-line computations. Unfortunately, they also typically require an unobstructed view of the object whose pose is being detected. The presence of occlusion and background clutter precludes the use of the normalizations that are typically applied and significantly alters the appearance of the object under detection. This work presents an algorithm that is based on applying eigendecomposition to a quadtree representation of the image dataset used to describe the appearance of an object. This allows decisions concerning the pose of an object to be based on only those portions of the image in which the algorithm has determined that the object is not occluded. The accuracy and computational efficiency of the proposed approach is evaluated on 16 different objects with up to 50% of the object being occluded and on images of ships in a dockyard.