Learning Spatial-Temporal Implicit Neural Representations for Event-Guided Video Super-Resolution

Our Paper

Please feel free to contact me if you have any issues. (ylu066"at"connect.hkust-gz.edu.cn)

YouTuBe Video

Training or Testing

Here is an example of training. The entry point for all training and testing is ‘main.py’.

sh options/CED/egisr-ceds-continue-4x-addchannnel-t4.sh

Dataset

Our dataset is at this link. If you are interested, please download it.

LINK: https://pan.baidu.com/s/1O2hHFZZat7SOBA-DB8G9pQ 
Code: cvpr

Because the data set takes up too much storage, I have compressed the dataset to reduce storage bandwidth and facilitate your download. Specifically, I compressed the events into a ProtoBuf. The description file of pb is tools/1-CompressDataset/event_frame.proto. For more information about the use of pb, please refer to the documentation (https://protobuf.dev). Read the pb file, please refer to the code tools/1-CompressDataset/read_pb.py During training, converting pb to numpy files will speed up training.

Citations

If this work is helpful for your research, please consider citing the following BibTeX entry.

@inproceedings{lu2023learning,
  title={Learning Spatial-Temporal Implicit Neural Representations for Event-Guided Video Super-Resolution},
  author={Lu, Yunfan and Wang, Zipeng and Liu, Minjie and Wang, Hongjian and Wang, Lin},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={1557--1567},
  year={2023}
}

Acknowledged items

Thanks to these two open source projects. I also hope my project can help you.

ESRT MMP-RNN