Video Visual Relation Detection Helpler

This repository contains some helper functions for the convenient usage of ImageNet-VidVRD dataset and VidOR dataset. It also contains scripts for evaluating several relevant tasks, i.e. video object detection and video relation detection.

Please note that the enclosed baseline does not serve as general baseline for those tasks. You can ignore it if you are not working with ImageNet-VidVRD dataset.

Please cite the following papers if the datasets help your research:

@inproceedings{shang2017video,
    author={Shang, Xindi and Ren, Tongwei and Guo, Jingfan and Zhang, Hanwang and Chua, Tat-Seng},
    title={Video Visual Relation Detection},
    booktitle={ACM International Conference on Multimedia},
    address={Mountain View, CA USA},
    month={October},
    year={2017}
}

@inproceedings{shang2019annotating,
    author={Shang, Xindi and Di, Donglin and Xiao, Junbin and Cao, Yu and Yang, Xun and Chua, Tat-Seng},
    title={Annotating Objects and Relations in User-Generated Videos},
    booktitle={ACM International Conference on Multimedia Retrieval},
    address={Ottawa, ON, Canada},
    month={June},
    year={2019}
}

Basic Prerequisites

• Python>=3.6
• numpy
• tqdm

Acknowledgement

This research is supported by the National Research Foundation, Singapore under its International Research Centres in Singapore Funding Initiative. Part of this research is also supported by National Science Foundation of China (61321491, 61202320), Collaborative Innovation Center of Novel Software Technology and Industrialization, and China Scholarship Council. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not reflect the views of National Research Foundation, Singapore.