Skip to content
No description, website, or topics provided.
Python
Branch: master
Clone or download
Latest commit a50ee2e Jul 27, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md Update README.md Jul 27, 2019
cvpr19.pdf Add files via upload Apr 27, 2019
full_version.pdf Add files via upload Apr 27, 2019
model.py Add files via upload Jul 27, 2019
test_demo.py Add files via upload Jul 27, 2019
test_video_01.zip Add files via upload Jul 27, 2019
test_video_02.zip Add files via upload Jul 27, 2019
test_video_03.zip Add files via upload Jul 27, 2019
test_video_04.zip Add files via upload Jul 27, 2019
utils.py Add files via upload Jul 27, 2019

README.md

Learning-to-Extract-Flawless-Slow-Motion-from-Blurry-Videos

This repository is a PyTorch implementation of the paper "Learning to Extract Flawless Slow Motion from Blurry Videos" from CVPR 2019 [paper][full version][video]

If you find our work useful in your research or publication, please cite our work:

@InProceedings{Jin_2019_CVPR,
author = {Jin, Meiguang and Hu, Zhe and Favaro, Paolo},
title = {Learning to Extract Flawless Slow Motion from Blurry Videos},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}

Requirements

This code has been tested with Python 3.7 and Pytorch 1.1.0.

Test

Unzip a real test video and download the pretrained model (cvpr19_model.pth) from google drive. This script will generate a 10x slow motion video.

unzip test_video_01.zip
python test_demo.py --cuda --model cvpr19_model.pth --input test_video_01 --out result

Training

To be updated

Contact

If you have any suggestions and questions, please send an email to jinmeiguang@gmail.com

You can’t perform that action at this time.