AcfNet

This repository contains the code (in PyTorch) for "Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching", accepted in AAAI 2020.

Notes

• We have provided all modules used in our AcfNet, and it can be directly integrated into our recent released DenseMatchingBenchmark for training.

Requirements:

• PyTorch1.1+, Python3.5+, Cuda10.0+

Reference:

If you find the code useful, please cite our paper:

@article{zhang2019adaptive,
  title={Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching},
  author={Zhang, Youmin and Chen, Yimin and Bai, Xiao and Zhou, Jun and Yu, Kun and Li, Zhiwei and Yang, Kuiyuan},
  journal={arXiv preprint arXiv:1909.03751},
  year={2019}
}