PWC Net
1.0.0
版权(C)2018 NVIDIA Corporation。版权所有。根据CC BY-NC-SA 4.0许可(https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode)获得许可。
对于Caffe用户,请参阅Caffe/readme.md。
有关Pytorch用户,请参考pytorch/readme.md
Pytorch的实施几乎与CAFFE实施相匹配(Sintel训练集的最终传球:Pytorch的2.31,Caffe的平均EPE)。
PWC-NET融合了几种经典的光流估计技术,包括图像金字塔,翘曲和成本量,以端到端可训练的深度神经网络,以实现最先进的结果。

Deqing Sun,Xiaodong Yang,Ming-Yu Liu和Jan Kautz。 “ PWC-NET:使用金字塔,翘曲和成本量的光流的CNN。” CVPR 2018或ARXIV:1709.02371
更新和扩展版本:“模型很重要,培训也是如此:CNN的经验研究,用于光流估计。” Arxiv:1809.05571
项目页面链接
在强大的视力挑战研讨会上谈论
在CVPR 2018会议上谈话
如果您使用PWC-NET,请引用以下论文:
@InProceedings{Sun2018PWC-Net,
author = {Deqing Sun and Xiaodong Yang and Ming-Yu Liu and Jan Kautz},
title = {{PWC-Net}: {CNNs} for Optical Flow Using Pyramid, Warping, and Cost Volume},
booktitle = CVPR,
year = {2018},
}
或arxiv纸
@article{sun2017pwc,
author={Sun, Deqing and Yang, Xiaodong and Liu, Ming-Yu and Kautz, Jan},
title={{PWC-Net}: {CNNs} for Optical Flow Using Pyramid, Warping, and Cost Volume},
journal={arXiv preprint arXiv:1709.02371},
year={2017}
}
或更新和扩展版本
@article{Sun2018:Model:Training:Flow,
author={Sun, Deqing and Yang, Xiaodong and Liu, Ming-Yu and Kautz, Jan},
title={Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimation},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
note = {to appear}
}
对于多帧流程,也请引用
@inproceedings{ren2018fusion,
title={A Fusion Approach for Multi-Frame Optical Flow Estimation},
author={Ren, Zhile and Gallo, Orazio and Sun, Deqing and Yang, Ming-Hsuan and Sudderth, Erik B and Kautz, Jan},
booktitle={Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV)},
year={2019}
}
Fownet2-Pytorch
通过移动摄像头进行3D运动场估计的动态场景中的刚性(ECCV 2018)
deqing sun([email protected])