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.

• PyTorch 1.7.1
• CUDA 11.0
• CuDNN 8.0.5
• python 3.6.7

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 [email protected]:ZhengChang467/STIPHR.git

    $ python STIP_run.py --dataset ucfsport

    $ python STIP_run.py --dataset human36m

    $ python STIP_run.py --dataset sjtu4k

We plan to share the training soon!

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}
}

See MIT License