Binocular Depth Estimation Based on Diffractive Optical Elements and the Semiglobal Matching Algorithm

Stereo vision is an important approach to obtain depth information from images. Widely known as one kind of passive approaches, conventional binocular stereo vision has been playing an important role in depth estimation field, especially in fine structure recovery. However, when it comes to places without obvious features, the accuracy of image matching is relatively lower. Taking that into consideration, a method combining the encoded structured light that can increase the number of features and binocular stereo vision is proposed. But the calibration of projector is difficult. In view of this, this paper presents a novel scheme by jointly using diffractive optical elements (DOE) to create speckle instead of encoded structured light and the semi-global matching (SGM) algorithm to improve the accuracy and efficiency of depth maps. Our experimental results showed that the proposed method can improve the quality of disparity maps comparable to the state of art stereo methods.

[1]  Adam L. Kaczmarek,et al.  Stereo vision with Equal Baseline Multiple Camera Set (EBMCS) for obtaining depth maps of plants , 2017, Comput. Electron. Agric..

[2]  Uwe Franke,et al.  Stereo vision during adverse weather - Using priors to increase robustness in real-time stereo vision , 2017, Image Vis. Comput..

[3]  Tieniu Tan,et al.  High quality depth map estimation of object surface from light-field images , 2017, Neurocomputing.

[4]  Zhongjie Zhu,et al.  Hybrid scheme for accurate stereo matching , 2017, Neurocomputing.

[5]  Haidi Ibrahim,et al.  Improvement of stereo matching algorithm for 3D surface reconstruction , 2018, Signal Process. Image Commun..

[6]  Xuming He,et al.  Learning to refine depth for robust stereo estimation , 2018, Pattern Recognit..

[7]  Byung-Gyu Kim,et al.  A local stereo matching algorithm based on weighted guided image filtering for improving the generation of depth range images , 2017, Displays.

[8]  Elmar Eisemann,et al.  Depth annotations: Designing depth of a single image for depth-based effects , 2018, Comput. Graph..

[9]  Shang-Hong Lai,et al.  Video synthesis from stereo videos with iterative depth refinement , 2017, J. Vis. Commun. Image Represent..

[10]  Abolghasem A. Raie,et al.  Dynamic programming-based dense stereo matching improvement using an efficient search space reduction technique , 2018 .

[11]  Zhiyu Zhou,et al.  Stereo matching using dynamic programming based on differential smoothing , 2016 .

[12]  Vladimir Kolmogorov,et al.  Visual correspondence using energy minimization and mutual information , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[13]  Tarkan Aydin,et al.  Stereo depth estimation using synchronous optimization with segment based regularization , 2010, Pattern Recognit. Lett..