epiception / SGM-Census

Semiglobal Matching with Census Matching Cost

SGM-Census

Simple CPU implementation of Semi-Global Matching using Census Transform and Hamming Distance Matching Cost.

---

Instructions

$git clone https://github.com/epiception/SGM-Census.git
$cd ~/path/to/SGM-Census
$make

Usage

./sgm <right image> <left image> <output image file> <disparity range>

Examples

./sgm Examples/teddy/right.png Examples/teddy/left.png output_disparity.png 64

---

Resources

• Accurate and Efficient Stereo Processing by Semi-Global Matching and Mutual Information
• Semi-Global Matching - Motivation, Developments and Applications
• Semi-Global Matching
• Middlebury Stereo Datasets
• Kitti Stereo Dataset

References

• Main Reference: https://github.com/reisub/SemiGlobal-Matching
• Terminal progress bars: https://github.com/luigipertoldi/progressbar