Yuchen Liu
ABSTRACT
We propose a real-time dual-window-based dense Stereo Correspondence method. Although there have been a large number of related research in decades, Compared with deep learning and global optimization algorithms, correlation window based methods still have the advantages of speed, low memory requirements, and high robustness. The challenge in window-based method is to choose a matching window size that maintains detail and objects’ boundaries at the same time coverage of areas with low or repeated textures. In the face of this trade-off, the classical method prefers to choose a larger window in order to achieve higher robustness.
We tackle this issue from a novel direction by proposing a Fast-Marching sampling algorithm to capture global features from a large window. Then aggregating the result with a small window with higher performance on details and objects’ boundaries to generate costs and perform WTA through a proposed continuous energy function. To verify the performance and feasibility of our method, Middlebury’s Stereo Matching benchmark is used. The experimental results demonstrated that our method can not only capture well the non-textured areas but also the details in the scene.
Supplemental Material
Available for Download
- Jia-Ren Chang and Yong-Sheng Chen. 2018. Pyramid Stereo Matching Network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Google Scholar
Cross Ref
- Anette Eltner and Danilo Schneider. 2015. Analysis of different methods for 3D reconstruction of natural surfaces from parallel-axes UAV images. The Photogrammetric Record 30, 151 (2015), 279–299.Google ScholarCross Ref
- A. Fusiello, V. Roberto, and E. Trucco. 1997. Efficient stereo with multiple windowing. In Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 858–863. https://doi.org/10.1109/CVPR.1997.609428Google ScholarCross Ref
- Andreas Geiger, Martin Roser, and Raquel Urtasun. 2011. Efficient large-scale stereo matching. In Computer Vision–ACCV 2010: 10th Asian Conference on Computer Vision, Queenstown, New Zealand, November 8-12, 2010, Revised Selected Papers, Part I 10. Springer, 25–38.Google Scholar
- Giorgio Grisetti, Rainer Kümmerle, Cyrill Stachniss, and Wolfram Burgard. 2010. A Tutorial on Graph-Based SLAM. IEEE Intelligent Transportation Systems Magazine 2, 4 (2010), 31–43. https://doi.org/10.1109/MITS.2010.939925Google ScholarCross Ref
- Raj Kumar Gupta and Siu-Yeung Cho. 2013. Window-based approach for fast stereo correspondence. IET Computer Vision 7, 2 (2013), 123–134. https://doi.org/10.1049/iet-cvi.2011.0077 arXiv:https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-cvi.2011.0077Google ScholarDigital Library
- Rostam Affendi Hamzah, Haidi Ibrahim, and Anwar Hasni Abu Hassan. 2017. Stereo matching algorithm based on per pixel difference adjustment, iterative guided filter and graph segmentation. Journal of Visual Communication and Image Representation 42 (2017), 145–160. https://doi.org/10.1016/j.jvcir.2016.11.016Google ScholarDigital Library
- Rostam Affendi Hamzah, A. Fauzan Kadmin, M. Saad Hamid, S. Fakhar A. Ghani, and Haidi Ibrahim. 2018. Improvement of stereo matching algorithm for 3D surface reconstruction. Signal Processing: Image Communication 65 (2018), 165–172. https://doi.org/10.1016/j.image.2018.04.001Google ScholarCross Ref
- Heiko Hirschmuller. 2007. Stereo processing by semiglobal matching and mutual information. IEEE Transactions on pattern analysis and machine intelligence 30, 2 (2007), 328–341.Google ScholarDigital Library
- Heiko Hirschmüller, Peter R Innocent, and Jon Garibaldi. 2002. Real-time correlation-based stereo vision with reduced border errors. International Journal of Computer Vision 47 (2002), 229–246.Google ScholarDigital Library
- Hiroshi Ishikawa and Davi Geiger. 1998. Occlusions, discontinuities, and epipolar lines in stereo. In Computer Vision—ECCV’98: 5th European Conference on Computer Vision Freiburg, Germany, June, 2–6, 1998 Proceedings, Volume I 5. Springer, 232–248.Google Scholar
- M. Kalia, N. Navab, and T. Salcudean. 2019. A Real-Time Interactive Augmented Reality Depth Estimation Technique for Surgical Robotics. In 2019 International Conference on Robotics and Automation (ICRA). 8291–8297. https://doi.org/10.1109/ICRA.2019.8793610Google ScholarDigital Library
- Takeo Kanade and Masatoshi Okutomi. 1994. A stereo matching algorithm with an adaptive window: Theory and experiment. IEEE transactions on pattern analysis and machine intelligence 16, 9 (1994), 920–932.Google Scholar
- Qian Long, Qiwei Xie, Seiichi Mita, Kazuhisa Ishimaru, and Noriaki Shirai. 2014. A real-time dense stereo matching method for critical environment sensing in autonomous driving. In 17th International IEEE Conference on Intelligent Transportation Systems (ITSC). 853–860. https://doi.org/10.1109/ITSC.2014.6957796Google ScholarCross Ref
- Faisal Mahmood and Nicholas J Durr. 2018. Deep learning and conditional random fields-based depth estimation and topographical reconstruction from conventional endoscopy. Medical image analysis 48 (2018), 230–243.Google Scholar
- Daniel Scharstein and Richard Szeliski. 2002. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. International journal of computer vision 47 (2002), 7–42.Google Scholar
- Olga Veksler. 2001. Stereo matching by compact windows via minimum ratio cycle. In Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001, Vol. 1. IEEE, 540–547.Google ScholarCross Ref
- O. Veksler. 2003. Fast variable window for stereo correspondence using integral images. In 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings., Vol. 1. I–I. https://doi.org/10.1109/CVPR.2003.1211403Google ScholarCross Ref
- Kang Zhang, Jiyang Li, Yijing Li, Weidong Hu, Lifeng Sun, and Shiqiang Yang. 2014. Binary Stereo Matching. arxiv:1402.2020 [cs.CV]Google Scholar
Index Terms
- A Fast Self-supervised Dual Correlation Window based Stereo Matching
Recommendations
Window‐based approach for fast stereo correspondence
In this study, the authors present a new area‐based stereo matching algorithm that computes dense disparity maps for a real‐time vision system. Although many stereo matching algorithms have been proposed in recent years, correlation‐based algorithms still ...
A Stereo Matching Algorithm with an Adaptive Window: Theory and Experiment
A central problem in stereo matching by computing correlation or sum of squared differences (SSD) lies in selecting an appropriate window size. The window size must be large enough to include enough intensity variation for reliable matching, but small ...
SUSAN window based cost calculation for fast stereo matching
CIS'05: Proceedings of the 2005 international conference on Computational Intelligence and Security - Volume Part IIThis paper presents a fast stereo matching algorithm using SUSAN window. The response of SUSAN window is used to calculate the dissimilarity cost. From this dissimilarity cost, an initial match can be found. Then, with this initial match, a dynamic ...
Comments