Fast Depth Estimation Method from a Pair of Retified Images

This paper proposes a line segment method to estimate the depth information from a pair of rectified images. The proposed method can achieve real-time and high quality stereo-matching. We first use a simple edge detection to find out the line segments in the reference image and then calculate the color difference of each line segment from a pair of rectified images. Finally, the minimum difference of each line segment is found out as the corresponding points. Unfortunately the line segments matching method is susceptible to cause error in occluded areas, so three methods are developed to refine the depth map. From the experiments, it is proved that the proposed method can accurately estimate the depth information in real-time.

[1]  D. Nistér,et al.  Stereo Matching with Color-Weighted Correlation, Hierarchical Belief Propagation, and Occlusion Handling , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  William T. Freeman,et al.  Comparison of graph cuts with belief propagation for stereo, using identical MRF parameters , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[3]  Takeo Kanade,et al.  Stereo by Intra- and Inter-Scanline Search Using Dynamic Programming , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Marc Pollefeys,et al.  Temporally Consistent Reconstruction from Multiple Video Streams Using Enhanced Belief Propagation , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[5]  Olga Veksler,et al.  Stereo correspondence by dynamic programming on a tree , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[6]  Alfred Schmitt,et al.  Real-Time Stereo by using Dynamic Programming , 2003, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[7]  Andreas Klaus,et al.  Segment-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[8]  Jian Sun,et al.  Symmetric stereo matching for occlusion handling , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[9]  Stefan Lüke,et al.  Real-Time Stereo Vision: Making More Out of Dynamic Programming , 2009, CAIP.

[10]  William T. Freeman,et al.  Understanding belief propagation and its generalizations , 2003 .

[11]  Yo-Sung Ho,et al.  Segment-Based Multi-View Depth Map Estimation Using Belief Propagation from Dense Multi-View Video , 2008, 2008 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video.

[12]  D. Scharstein,et al.  A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms , 2001, Proceedings IEEE Workshop on Stereo and Multi-Baseline Vision (SMBV 2001).

[13]  Antonios Gasteratos,et al.  Stereo vision for robotic applications in the presence of non-ideal lighting conditions , 2010, Image Vis. Comput..

[14]  Yi Deng,et al.  A Fast Line Segment Based Dense Stereo Algorithm Using Tree Dynamic Programming , 2006, ECCV.

[15]  Sang Uk Lee,et al.  A dense stereo matching using two-pass dynamic programming with generalized ground control points , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[16]  Ruigang Yang,et al.  Stereo Matching with Color-Weighted Correlation, Hierarchical Belief Propagation and Occlusion Handling , 2006, CVPR.

[17]  Miao Liao,et al.  High-Quality Real-Time Stereo Using Adaptive Cost Aggregation and Dynamic Programming , 2006, Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06).

[18]  Minglun Gong,et al.  Fast stereo matching using reliability-based dynamic programming and consistency constraints , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.