README.md

# csl-openworld Dataset Release Agreement

## Non-Exclusive License Agreement

This Non-Exclusive License Agreement (the "Agreement") is made between **Intelligent Sensing and Visual Computing / Nanjing University** (the "Provider") and:

____________________________________________________________

(Please fill in the name of your research group, laboratory, university, institution, or organization on the line above.)

By downloading, accessing, receiving, or using the **csl-openworld Dataset** (the "Dataset"), the licensee and its researchers (collectively, the "Licensee") acknowledge that they have read, understood, and agreed to comply with the terms and conditions of this Agreement.

## 1. Dataset Description and Copyright Notice

The csl-openworld Dataset is collected, processed, annotated, and curated by the Provider for the purpose of supporting research in sign language understanding, including but not limited to sign language recognition, translation, retrieval, evaluation, and related machine learning tasks.

The Dataset may include video clips, video frames, annotations, labels, metadata, derived files, and other related materials.

The Provider does not claim ownership of the copyright or other intellectual property rights in original raw video content that may have been collected from publicly accessible sources. Such original content remains the property of its respective copyright holders. The annotations, labels, processing, organization, selection, structure, and compilation of the Dataset are the intellectual property of the Provider, unless otherwise stated.

The Dataset is released for **research purposes only**.

## 2. License Grant

Subject to the terms of this Agreement, the Provider grants the Licensee a limited, non-exclusive, non-transferable, revocable license to use the Dataset solely for academic, non-profit, and scientific research purposes.

No ownership rights are transferred to the Licensee under this Agreement.

## 3. Research-Only and Non-Commercial Use

The Licensee agrees that the Dataset, in whole or in part, shall be used only for academic research, non-profit scientific studies, benchmarking, evaluation, and preparation of academic publications, theses, dissertations, reports, or presentations.

The Dataset shall not be used for any commercial purpose in any form, including but not limited to:

- Sales, marketing, advertising, or product promotion;
- Product development or commercial service deployment;
- Training, fine-tuning, evaluating, or improving models, algorithms, systems, or technologies that are used or intended to be used for commercial purposes;
- Incorporation into any commercial product, paid service, proprietary system, or commercial dataset.

## 4. Restrictions on Publication and Redistribution

The Licensee shall not redistribute, publish, copy, upload, sublicense, transfer, sell, lease, rent, or further disseminate the Dataset, in whole or in part, in any way or form, whether for profit or not.

This restriction includes, but is not limited to, making the Dataset available on public websites, repositories, cloud storage services, model hubs, file-sharing platforms, or distributing it to another laboratory, research group, facility, organizational unit, university, institution, company, or third party.

The Licensee shall not make or permit anyone to make copies of the whole or any part of the Dataset except as strictly necessary for the approved research purpose and internal use by authorized researchers under this Agreement.

Any collaborator from another institution or organization must obtain separate authorization from the Provider unless prior written permission is granted by the Provider.

## 5. Privacy, Ethics, and Responsible Use

The Licensee shall use the Dataset in a lawful, ethical, and socially responsible manner.

The Licensee shall not use the Dataset to identify, re-identify, contact, track, profile, or otherwise target individuals appearing in the Dataset, except where such individuals are public figures already explicitly identified in the original public source.

The Licensee shall avoid using or publishing any part of the Dataset where the identity of individuals is readily recognizable, either as a stand-alone extract or when combined with other materials, unless such use is necessary for the approved research purpose and permitted by applicable law, institutional policy, and ethical requirements.

The Licensee shall not use the Dataset to generate defamatory, discriminatory, misleading, harmful, offensive, unlawful, or unethical content.

Where required by the Licensee's institution or applicable regulations, the Licensee shall obtain appropriate ethics approval, institutional review board approval, or equivalent authorization before using the Dataset.

## 6. Compliance and Review

The Licensee is responsible for ensuring that its use of the Dataset complies with applicable laws, regulations, institutional policies, platform terms, copyright rules, privacy requirements, and ethical standards.

Upon reasonable request from the Provider, the Licensee shall provide information sufficient for the Provider to review whether the Dataset is being used in accordance with this Agreement.

Failure to comply with this Agreement may result in termination of access to the Dataset, denial of access to future versions of the Dataset, and possible legal liability.

## 7. Dataset Access and Download

To request access to the Dataset, the Licensee must:

1. Download this Agreement.
2. Read and sign this Agreement.
3. Send the signed Agreement to **yafeng@nju.edu.cn** by email.
4. After review and approval by the Provider, the Licensee will receive a download link for the Dataset by email.

The Provider reserves the right to approve or deny access requests at its discretion.

## 8. Citation

If the Licensee uses the Dataset in any paper, preprint, report, thesis, dissertation, presentation, or other academic work, the Licensee shall cite the following publication:

```bibtex
@article{gan2026onesign,
  title={OneSign: Unifying Sign Language Understanding Tasks with One Model},
  author={Gan, Shiwei and Yin, Yafeng and Liu, Xiao and Tuerdaken, Desibieer and Jiang, Zhiwei and Xie, Lei and Lu, Sanglu},
  year={2026},
  publisher={arXiv}
}
```

The Licensee should also acknowledge the Dataset as follows:

> This work uses the csl-openworld Dataset collected and curated by Intelligent Sensing and Visual Computing / Nanjing University.

The Licensee shall not imply that the Provider endorses the Licensee's research, results, products, services, or conclusions unless prior written permission has been obtained from the Provider.

## 9. Dataset Updates, Removal, and Takedown

The Provider may update, modify, restrict, suspend, or discontinue access to the Dataset at any time.

If the Provider receives a valid copyright, privacy, ethical, or legal complaint concerning any part of the Dataset, the Provider may remove or modify the relevant data. Upon notification, the Licensee agrees to promptly delete or stop using any removed or restricted portion of the Dataset.

## 10. Termination

This Agreement is effective upon the Licensee's access to the Dataset and remains in effect until terminated.

The Provider may terminate the Licensee's access immediately if the Licensee violates any term of this Agreement. Upon termination, the Licensee must cease all use of the Dataset and delete all copies in its possession or control, including backup copies, unless retention is required by law or institutional policy.

Sections concerning attribution, disclaimers, limitation of liability, indemnification, and third-party rights shall survive termination.

## 11. Disclaimer of Warranties

The Dataset is provided "as is" and "as available", without warranty of any kind, whether express, implied, statutory, or otherwise.

The Provider makes no warranties regarding the accuracy, completeness, reliability, availability, fitness for a particular purpose, non-infringement, or legal usability of the Dataset.

The Provider does not guarantee that the Dataset is free from errors, omissions, offensive content, bias, privacy concerns, or third-party rights claims.

## 12. Limitation of Liability

To the maximum extent permitted by applicable law, the Provider shall not be liable for any direct, indirect, incidental, consequential, special, exemplary, or punitive damages arising out of or related to the Licensee's access to or use of the Dataset.

The Licensee assumes all risks associated with the use of the Dataset.

## 13. Indemnification

The Licensee agrees to indemnify, defend, and hold harmless the Provider, its researchers, employees, students, affiliates, and institutions from and against any claims, damages, losses, liabilities, costs, or expenses arising from:

- The Licensee's use or misuse of the Dataset;
- The Licensee's violation of this Agreement;
- The Licensee's violation of applicable laws, regulations, ethical requirements, or third-party rights;
- Any copyright, privacy, publicity, defamation, or data protection claim related to the Licensee's use of the Dataset.

## 14. Final Interpretation

The Provider reserves the final right to interpret this Agreement.

## 15. Contact

For questions, permission requests, access requests, or takedown notices related to the Dataset, please contact:

**Provider:** Intelligent Sensing and Visual Computing / Nanjing University  
**Email:** yafeng@nju.edu.cn

---

## User Acknowledgement

By signing below, I acknowledge that I have read, understood, and agreed to the terms of this Agreement.

**Name:** ________________________________________________

**Email:** ________________________________________________

**Institute / Organization:** _______________________________

**Research Group / Laboratory:** ___________________________

**Position / Title:** ______________________________________

**Address:** _____________________________________________

**Research Purpose:** ____________________________________

**Signature:** ____________________________________________

**Date:** ________________________________________________