Skip to content

Latest commit

 

History

History
50 lines (43 loc) · 3.92 KB

LICENSE.md

File metadata and controls

50 lines (43 loc) · 3.92 KB
Raw

DIVOTrack: A Cross-View Dataset for Multi-Human Tracking in DIVerse Open Scenes

The DIVOTrack dataset was created by Shenghao Hao, Peiyuan Liu, Yibing Zhan, Kaixun Jin, Zuozhu Liu, Mingli Song, Jenq-Neng Hwang, Gaoang Wang. DIVOTrack is a novel cross-view multi-human tracking dataset which is more realistic and diverse, has more tracks, and incorporates moving cameras. Accordingly, a standardized benchmark is built for cross-view tracking, with clear split of training and testing set, publicly accessible detection, and standard cross-view tracking evaluation metrics. With the proposed dataset and benchmark, the cross-view tracking methods can be fairly compared in the future, which will improve the development of cross-view tracking techniques. now made publicly available for research purpose only on a case-by-case basis. DIVOTrack is serving as the technical agent for distribution of the dataset and reserves the copyright of all the images in the database. Researchers who request the DIVOTrack dataset must sign this agreement and thereby agree to obey the restrictions listed in this document. Failure to observe the restrictions might result in access being denied for the request of the future version of the DIVOTrack dataset and being subject to civil damages in the case of publication of images that have not been approved for release. The researchers agree to the following restrictions on using DIVOTrack dataset:

  1. All the videos and images will be published or released in technical reports, or papers only. Any videos or images will never be released in commercial materials, newspapers, or other public media.
  2. The dataset will not be re-distributed, published, copied, or further disseminated in any way or form whatsoever, whether for profit or not. This includes further distributing, copying, or disseminating to a different facility or organizational unit in the requesting university, organization, or company.
  3. All the data in DIVOTrack dataset can only be used for the purpose of scientific researches. The DIVOTrack, in whole or in part, will not be used for any commercial, military and other illegal purpose in any form.
  4. All technical papers, documents and reports which used the DIVOTrack dataset should acknowledge the use of the database as follows: "The research in this paper used the DIVOTrack dataset collected by Zhejiang University." and provide citations to the corresponding article:
@article{hao2024divotrack,
  title={Divotrack: A novel dataset and baseline method for cross-view multi-object tracking in diverse open scenes},
  author={Hao, Shengyu and Liu, Peiyuan and Zhan, Yibing and Jin, Kaixun and Liu, Zuozhu and Song, Mingli and Hwang, Jenq-Neng and Wang, Gaoang},
  journal={International Journal of Computer Vision},
  volume={132},
  number={4},
  pages={1075--1090},
  year={2024},
  publisher={Springer}
}
  1. A student applicant should ask his/her supervisor to sign this agreement, and send the signed agreement to us using a formal university email (do not use emails like xxx@163.com, xxx@gmail.com; otherwise the application will not be considered). A student applicant should also copy the application email to his/her supervisor while sending the agreement to us.
  2. Please send the application with signed agreement to Shengyu Hao (shengyuhao@zju.edu.cn) and Gaoang Wang (gaoangwang@intl.zju.edu.cn). Again, a student applicant should also copy the application email to his/her supervisor while sending the agreement to us.
  3. The final interpretation right of this agreement belongs to Zhejiang University and authors.

(For a student applicant, please provide your supervisor’s information below and ask your supervisor to sign the agreement)

Faculty Name:_____________ Signature:_____________
Email:_____________ Faculty Homepage:__________________
University and Department:__________________
Date:_________ (MM/DD/YY)