pytorch / pytorch/tutorials

copyright and permission question about celebA dataset used in tutorials

Open
#840 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
9.3k
Forks
4.4k
Avg merge
1d 21h
Merged PRs (30d)
4

Description

I have a question about the legality/ethics around the use of the celebA dataset.

The data is scrapped from the public web, and packaged and redistribute day the Chinese University in Hong Kong, the effective upstream source of the celebA dataset.

They state they don't own the data and also require the dataset is not used for commercial use.

However some questions remain which are relevant in jurisdictions like the US, UK and much of Europe:

  • The photos are personal data, and I can't find any source which states that the data subjects have given permission for their data to be used in ways other than the original (eg celebrity news websites). Use on a tutorial site such as this is a different use beyond the original agreed to.

  • There may be ownership / rights / permissions related to photographic and news media agencies. I have found no source to state that permission has been granted.

I hope I am unduly concerned here, please do correct my rationale if I am mistaken.

My own specific interest is that I would like to use CelebA in a tutorial book. I have received specific permission to use the dataset from the Chinese University in Hong Kong, but further questions such as the above are as yet unanswered.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are mentioned. The issue asks for legal and ethical clarification about using the CelebA dataset in tutorials, so done would require an authoritative response or documented resolution rather than a code change.

Written by the indexing model from the issue text.

Assessment

Domain
documentation
Issue type
Documentation
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.