Awesome
neural-global-shutter
Neural Global Shutter: Learn to Restore Video from a Rolling Shutter Camera with Global Reset Feature
CVPR, 2022
paper | project website | dataset
Introduction
<div align="center"> <img src="resources/RSGR_problem.png" width="800"/> </div>Abbreviation: RS→rolling shutter; GS→global shutter; RSGR→rolling shutter with global reset
- We offer an alternative and new solution for RS rectification by turning the rectification problem into a deblur-like one with RSGR, a widely ignored feature of RS sensors.
- We propse a method to restore clean video from RSGR inputs by eliminating the spatial-varying distortions.
- We collect a dataset with paired RSGR and GS videos under real scenes for developing and evaluating algorithms.
Updates
- 2022/03/11: We make the RSGR-GS dataset public available.
Citation
If you find this project useful in your research, please consider cite:
@inproceedings{Wang_2022_CVPR,
author = {Wang, Zhixiang and Ji, Xiang and Huang, Jia-Bin and Satoh, Shin'ichi and Zhou, Xiao and Zheng, Yinqiang},
title = {Neural Global Shutter: Learn to Restore Video from a Rolling Shutter Camera with Global Reset Feature},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2022}
}
Contact
If you have any questions, please contact wangzx1994 -a-t- gmail.com or yqzheng -a-t ai.u-tokyo.ac.jp
<!-- ## FAQ TBA -->