Joint-task Self-supervised Learning for Temporal Correspondence

Project | Paper

Overview

Joint-task Self-supervised Learning for Temporal Correspondence

Xueting Li*, Sifei Liu*, Shalini De Mello, Xiaolong Wang, Jan Kautz, Ming-Hsuan Yang.

(* equal contributions)

In Neural Information Processing Systems (NeurIPS), 2019.

Citation

If you use our code in your research, please use the following BibTex:

@inproceedings{uvc_2019,
    Author = {Xueting Li and Sifei Liu and Shalini De Mello and Xiaolong Wang and Jan Kautz and Ming-Hsuan Yang},
    Title = {Joint-task Self-supervised Learning for Temporal Correspondence},
    Booktitle = {NeurIPS},
    Year = {2019},
}

Instance segmentation propagation on DAVIS2017

Prerequisites

The code is tested in the following environment:

• Ubuntu 16.04
• Pytorch 1.1.0, tqdm, scipy 1.2.1

Testing on DAVIS2017

Testing without tracking

To test on DAVIS2017 for instance segmentation mask propagation, please run:

python test.py -d /workspace/DAVIS/ -s 480

Important parameters:

• -c: checkpoint path.
• -o: results path.
• -d: DAVIS 2017 dataset path.
• -s: test resolution, all results in the paper are tested on 480p images, i.e. -s 480.

Please check the test.py file for other parameters.

Testing with tracking

To test on DAVIS2017 by tracking & propagation, please run:

python test_with_track.py -d /workspace/DAVIS/ -s 480

Similar parameters as test.py, please see the test_with_track.py for details.

Training on Kinetics

Dataset

We use the kinetics dataset for training.

Training command

python track_match_v1.py --wepoch 10 --nepoch 30 -c match_track_switch --batchsize 40 --coord_switch 0 --lc 0.3

Acknowledgements

• This code is based on TPN and TimeCycle.
• For any issues, please contact [email protected] or [email protected].