A naïve implementation of the Fast Bilateral-Space Stereo paper by Jonathan T. Barron. [1] [2] [3]
The goal of this project was to get an intuition of this method. Therefore writing performant code was not a criteria.
Only the simplified bilateral grid method without the multiscale optimization is implemented.
The algorithm needs a stereo pair as input and will generate a disparity map.
More details about this implementation can be found at http://tvdz.be/2017/03/fast-bilateral-space-stereo/.
The following dependencies are needed.
The version numbers are the ones used during development.
- OpenCV 3.2.0
- (optional) opencv_contrib (if you want to use the domain transform filter to smoothen the disparity map)
- Eigen 3.3.2
- Ceres Solver 1.12.0
- Glog 0.3.3
- GFlags 2.2.0
The code was developed on a Windows machine with Visual Studio 2015.
[1] Barron, Jonathan T., et al. "Fast bilateral-space stereo for synthetic defocus." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015.
[2] Barron, Jonathan T., et al. "Fast bilateral-space stereo for synthetic defocus—Supplemental material." Proc. IEEE Conf. Comput. Vis. Pattern Recognit.(CVPR). 2015.
[3] Barron, Jonathan T., http://jonbarron.info/