Ndhlovu, TNicolls, F2009-12-032009-12-032009-11Ndhlovu, T and Nicolls, F. 2009. Alternative confidence measure for local matching stereo algorithms. 3rd Robotics and Mechatronics Symposium (ROBMECH 2009). Pretoria, South Africa, 8-10 November 2009, pp 59780620447218http://hdl.handle.net/10204/37913rd Robotics and Mechatronics Symposium (ROBMECH 2009). Pretoria, South Africa, 8-10 November 2009The authors present a confidence measure applied to individual disparity estimates in local matching stereo correspondence algorithms. It aims at identifying textureless areas, where most local matching algorithms fail. The confidence measure works by analyzing the correlation curve produced during the matching process. The authors also test the confidence measure by developing an easily parallelized local matching algorithm, and use our confidence measure to filter out unreliable disparity estimates. Using the Middlebury dataset and our own evaluation scheme, the results show that the confidence measure significantly decreases the disparity estimate errors at a low computational overhead.enConfidence measureStereo algorithmsRoboticsConference PresentationNdhlovu, T., & Nicolls, F. (2009). Alternative confidence measure for local matching stereo algorithms. http://hdl.handle.net/10204/3791Ndhlovu, T, and F Nicolls. "Alternative confidence measure for local matching stereo algorithms." (2009): http://hdl.handle.net/10204/3791Ndhlovu T, Nicolls F, Alternative confidence measure for local matching stereo algorithms; 2009. http://hdl.handle.net/10204/3791 .TY - Conference Presentation AU - Ndhlovu, T AU - Nicolls, F AB - The authors present a confidence measure applied to individual disparity estimates in local matching stereo correspondence algorithms. It aims at identifying textureless areas, where most local matching algorithms fail. The confidence measure works by analyzing the correlation curve produced during the matching process. The authors also test the confidence measure by developing an easily parallelized local matching algorithm, and use our confidence measure to filter out unreliable disparity estimates. Using the Middlebury dataset and our own evaluation scheme, the results show that the confidence measure significantly decreases the disparity estimate errors at a low computational overhead. DA - 2009-11 DB - ResearchSpace DP - CSIR KW - Confidence measure KW - Stereo algorithms KW - Robotics LK - https://researchspace.csir.co.za PY - 2009 SM - 9780620447218 T1 - Alternative confidence measure for local matching stereo algorithms TI - Alternative confidence measure for local matching stereo algorithms UR - http://hdl.handle.net/10204/3791 ER -