Stereo Matching

This project aims to estimate a disparity map, given two stereo images.

An on-line evaluation of current stereo matching algorithms: http://vision.middlebury.edu/stereo/, with many stereo datasets and stereo correspondance software.

A pdf simply introducing stereo vision: 04stereo.pdf

A ppt simply introducing stereo vision: 6-Stereo.ppt

References

[1] D. Scharstein and R. Szeliski. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms.
International Journal of Computer Vision, 47(1/2/3):7-42, April-June 2002.

[2] H. Hirschmüller and D. Scharstein. Evaluation of cost functions for stereo matching.
In IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2007), Minneapolis, MN, June 2007.

Assignment:

[1] Implement kinds of local methods, such as SSD, SAD, NCC. Evaluate the results by on-line evaluation.

[2] Implement global methods, using graph cuts and belief propagation. Evaluate the results by on-line evaluation.

[3]* Accelerate the algorithm using GPU. Graphcut GPU实现参考代码，Belief Propagation GPU实现参考文章。