Home

Awesome

STIP (extended from our previous work in CVPR2022)

Zheng Chang, Xinfeng Zhang, Shanshe Wang, Siwei Ma, Wen Gao.

Official PyTorch Code for "STIP: A Spatiotemporal Information-Preserving and Perception-Augmented Model for High-Resolution Video Prediction"

This work is extended from our previous work STRPM, which has been accepted by CVPR2022. The codes for STRPM have also been made public.

Requirements

Installation

Create conda environment:

    $ conda create -n STIP python=3.6.7
    $ conda activate STIP
    $ pip install -r requirements.txt
    $ conda install pytorch==1.7 torchvision cudatoolkit=11.0 -c pytorch

Download repository:

    $ git clone git@github.com:ZhengChang467/STIPHR.git

Test on the ucfsports dataset

    $ python STIP_run.py --dataset ucfsport

Test on the Human3.6M dataset

    $ python STIP_run.py --dataset human36m

Test on the SJTU4K dataset

    $ python STIP_run.py --dataset sjtu4k

We plan to share the training soon!

Citation

Please cite the following paper if you feel this repository useful.

@article{chang2022strpm,
title={STRPM: A Spatiotemporal Residual Predictive Model for High-Resolution Video Prediction},
author={Chang, Zheng and Zhang, Xinfeng and Wang, Shanshe and Ma, Siwei and Gao, Wen},
journal={arXiv preprint arXiv:2203.16084},
year={2022}
}

License

See MIT License