This is a personal reimplementation of SPyNet [1] using PyTorch. Should you be making use of this work, please cite the paper accordingly. Also, make sure to adhere to the licensing terms of the authors. Should you be making use of this particular implementation, please acknowledge it appropriately [2].
For the original Torch version of this work, please see: https://github.com/anuragranj/spynet
Another optical flow implementation from me: https://github.com/sniklaus/pytorch-pwc
And another optical flow implementation from me: https://github.com/sniklaus/pytorch-liteflownet
Yet another optical flow implementation from me: https://github.com/sniklaus/pytorch-unflow
To download the pre-trained models, run bash download.bash. These originate from the original authors, I just converted them to PyTorch.
To run it on your own pair of images, use the following command. You can choose between various models, please make sure to see their paper / the code for more details.
python run.py --model sintel-final --first ./images/first.png --second ./images/second.png --out ./out.flo
python run.py --model sintel-final --first ../VID_20191225_170802/frame0077.jpg --second ../VID_20191225_170802/frame0080.jpg --out ./out.flo
python compute_flow.py --flowfile out.flo --write True
I am afraid that I cannot guarantee that this reimplementation is correct. However, it produced results identical to the implementation of the original authors in the examples that I tried. Please feel free to contribute to this repository by submitting issues and pull requests.
As stated in the licensing terms of the authors of the paper, the models are free for non-commercial and scientific research purpose. Please make sure to further consult their licensing terms.
[1] @inproceedings{Ranjan_CVPR_2017,
author = {Ranjan, Anurag and Black, Michael J.},
title = {Optical Flow Estimation Using a Spatial Pyramid Network},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition},
year = {2017}
}
[2] @misc{pytorch-spynet,
author = {Simon Niklaus},
title = {A Reimplementation of {SPyNet} Using {PyTorch}},
year = {2018},
howpublished = {\url{https://github.com/sniklaus/pytorch-spynet}}
}