Low memory cost block-based belief propagation for stereo correspondence

Yu Cheng Tseng*, Nelson Chang, Tian-Sheuan Chang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

22 Scopus citations

Abstract

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 32×32 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.

Original languageEnglish
Title of host publicationProceedings of the 2007 IEEE International Conference on Multimedia and Expo, ICME 2007
Pages1415-1418
Number of pages4
DOIs
StatePublished - 1 Dec 2007
EventIEEE International Conference onMultimedia and Expo, ICME 2007 - Beijing, China
Duration: 2 Jul 20075 Jul 2007

Publication series

NameProceedings of the 2007 IEEE International Conference on Multimedia and Expo, ICME 2007

Conference

ConferenceIEEE International Conference onMultimedia and Expo, ICME 2007
CountryChina
CityBeijing
Period2/07/075/07/07

Fingerprint Dive into the research topics of 'Low memory cost block-based belief propagation for stereo correspondence'. Together they form a unique fingerprint.

  • Cite this

    Tseng, Y. C., Chang, N., & Chang, T-S. (2007). Low memory cost block-based belief propagation for stereo correspondence. In Proceedings of the 2007 IEEE International Conference on Multimedia and Expo, ICME 2007 (pp. 1415-1418). [4284925] (Proceedings of the 2007 IEEE International Conference on Multimedia and Expo, ICME 2007). https://doi.org/10.1109/ICME.2007.4284925