High-Quality Real-Time Stereo Using Adaptive Cost Aggregation and Dynamic Programming
- 1 June 2006
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 798-805
- https://doi.org/10.1109/3dpvt.2006.75
Abstract
We present a stereo algorithm that achieves high quality results while maintaining real-time performance. The key idea is simple: we introduce an adaptive aggregation step in a dynamic-programming (DP) stereo framework. The per-pixel matching cost is aggregated in the vertical direction only. Compared to traditional DP, our approach reduces the typical "streaking" artifacts without the penalty of blurry object boundaries. Evaluation using the benchmark Middlebury stereo database shows that our approach is among the best (ranked first in the new evaluation system) for DP-based approaches. The performance gain mainly comes from a computationally expensive weighting scheme based on color and distance proximity. We utilize the vector processing capability and parallelism in commodity graphics hardware to speed up this process over two orders of magnitude. Over 50 million disparity evaluations per second (MDE/s)1 are achieved in our current implementation.Keywords
This publication has 14 references indexed in Scilit:
- Real-Time Connectivity Constrained Depth Map Computation Using Programmable Graphics HardwarePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Video-rate stereo depth measurement on programmable hardwarePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A binocular stereo algorithm for reconstructing sloping, creased, and broken surfaces in the presence of half-occlusionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Improvements in real-time correlation-based stereo visionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Trinocular Stereo: A Real-Time Algorithm and its EvaluationInternational Journal of Computer Vision, 2002
- Real-Time Correlation-Based Stereo Vision with Reduced Border ErrorsInternational Journal of Computer Vision, 2002
- Depth Discontinuities by Pixel-to-Pixel StereoInternational Journal of Computer Vision, 1999
- A Maximum Likelihood Stereo AlgorithmComputer Vision and Image Understanding, 1996
- A space-sweep approach to true multi-image matchingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1996
- Stereo by Intra- and Inter-Scanline Search Using Dynamic ProgrammingIeee Transactions On Pattern Analysis and Machine Intelligence, 1985