Meaningful Matches in Stereovision
Abstract
This paper introduces a statistical method to decide whether two blocks in a pair of images match reliably. The method ensures that the selected block matches are unlikely to have occurred "just by chance." The new approach is based on the definition of a simple but faithful statistical background model for image blocks learned from the image itself. A theorem guarantees that under this model, not more than a fixed number of wrong matches occurs (on average) for the whole image. This fixed number (the number of false alarms) is the only method parameter. Furthermore, the number of false alarms associated with each match measures its reliability. This a contrario block-matching method, however, cannot rule out false matches due to the presence of periodic objects in the images. But it is successfully complemented by a parameterless self-similarity threshold. Experimental evidence shows that the proposed method also detects occlusions and incoherent motions due to vehicles and pedestrians in nonsimultaneous stereo.
Additional Information
© 2012 IEEE. Manuscript received 16 Aug. 2010; revised 25 July 2011; accepted 25 Sept. 2011; published online 8 Oct. 2011. Date of Current Version: 22 March 2012. The authors thank Pascal Getreuer for helpful comments on this work. This work was partially supported by the following projects: FREEDOM (ANR07-JCJC-0048-01), Callisto (ANR-09-CORD-003), ECOS Sud U06E01, and STIC Amsud (11STIC-01 - MMVPSCV). Neus Sabater was with ENS Cachan, CMLA, France, when the paper was written.Additional details
- Eprint ID
- 30095
- Resolver ID
- CaltechAUTHORS:20120416-094241649
- ANR07-JCJC-0048-01
- FREEDOM
- ANR-09-CORD-003
- Callisto
- U06E01
- ECOS Sud
- 11STIC-01 - MMVPSCV
- STIC Amsud
- Created
-
2012-04-17Created from EPrint's datestamp field
- Updated
-
2021-11-09Created from EPrint's last_modified field
- Other Numbering System Name
- INSPEC Accession Number
- Other Numbering System Identifier
- 12617613