Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation
University of Delaware, Newark, DE 19716
Workshop on Applications of Computer Vision (WACV), 2009, p. 1-8.
@conference{grauer2009hierarchical,
title={Hierarchical belief propagation to reduce search space using CUDA for stereo and motion estimation},
author={Grauer-Gray, S. and Kambhamettu, C.},
booktitle={Applications of Computer Vision (WACV), 2009 Workshop on},
pages={1–8},
issn={1550-5790},
year={2009},
organization={IEEE}
}
This paper describes a hierarchical belief propagation implementation in which a ‘rough’ disparity map calculation or motion estimation in higher levels is used to limit the search space and enable the calculation of the desired disparity map/set of motion vectors using a smaller search space than traditional belief propagation. We implement our algorithm on the GPU using the CUDA architecture and explore a number of implementation details with promising results; it is clear that the storage requirements of belief propagation can be significantly reduced using our method without too large of a sacrifice in the accuracy of the results. In addition, we take advantage of the interpolation capabilities built into the GPU in order to retrieve the computed disparities/motion vectors at sub-pixel accuracy without making any change in implementation.
March 22, 2011 by hgpu
