This is the PyTorch implementation of the paper "Efficient Deep Learning for Stereo Matching" by Luo, W. and Schwing, A. and Urtasun, R. done as part of Computer Vision summer school organized at IIIT Hyderabad, India.
For more details, refer the original torch implementation of the authors at https://bitbucket.org/saakuraa/cvpr16_stereo_public