標題: Low memory cost block-based belief propagation for stereo correspondence
作者: Tseng, Yu-Cheng
Chang, Nelson
Chang, Tian-Sheuan
電子工程學系及電子研究所
Department of Electronics Engineering and Institute of Electronics
公開日期: 2007
摘要: The typical belief propagation has good accuracy for stereo correspondence but suffers from large run-time memory cost. In this paper, we propose a block-based belief propagation algorithm for stereo correspondence that partitions an image into regular blocks for optimization. With independently partitioned blocks, the required memory size could be reduced significantly by 99% with slightly degraded performance with a 3202 block size when compared to original one. Besides, such blocks are also suitable for parallel hardware implementation. Experimental results using Middlebury stereo test bed demonstrate the performance of the proposed method.
URI: http://hdl.handle.net/11536/11246
ISBN: 978-1-4244-1016-3
期刊: 2007 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOLS 1-5
起始頁: 1415
結束頁: 1418
顯示於類別:會議論文