Skip to content

Dataset/LSA-T #88

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Jun 20, 2024
Merged

Conversation

cleong110
Copy link
Contributor

@cleong110 cleong110 commented Jun 19, 2024

Additional Checklist for datasets:

When adding a dataset, follow the following steps. This pull request provides an example:

  • Fork the repo
  • sync forks
  • git checkout master
  • git pull
  • New branch: dataset/something
  • Create a JSON along the lines of the schema below. e.g. FOO.json
  • Add the JSON to src/datasets. e.g. src/datasets/FOO.json
  • Add BibTex to src/references.bib.
    • prepend the citation key with dataset. e.g. dataset:sehyr2021asl
    • "language" field should not need "sign language". No need to say "American Sign Language", "American" will do.
    • Very concise "samples" field, the table does not have a lot of space to display it.
  • Commit/push the changes
  • Make a pull request!

Schema:

{
  "pub": {
    "name": string, # this gets used as the name of the dataset, e.g. "WLASL"
    "year": integer or null,
    "publication":string or null, # this matches a key in references.bib, e.g. "dataset:joshiISLTranslateDatasetTranslating2023"
    "url": string or null # URL to access it. e.g. "https://www.sign-lang.uni-hamburg.de/dgs-korpus/index.php/welcome.html"
  },
  "#loader": string or null, # the key you would use in the sign language datasets library. e.g. "dgs_corpus". Website will auto-link
  "#items": integer or null, # this is the number of unique signs in the column
  "#samples": string or null, # e.g. "1100 videos" or "8,257 Sentences"
  "#signers": integer or string or null, # number of unique signers
  "features": array of strings, ["feature1","feature2"], # I've seen things like "mouthing", "video:RGB", "pose:Kinect", "pose:OpenPose","text:Polish", "gloss:ASL", "writing:HamNoSys", etc.
  "language": string, # the Sign language or languages, e.g. "American" for American Sign Language (ASL)
  "license": string or null,
  "licenseUrl": string or null
}

Closes #32 if accepted

@cleong110 cleong110 marked this pull request as ready for review June 19, 2024 19:13
@cleong110 cleong110 changed the title Dataset/lsa t Dataset/LSA-T Jun 19, 2024
"#samples": "14,880 sentence-level videos",
"#signers": 103,
"license": "Public (MIT)",
"licenseUrl": "https://github.com/midusi/LSA-T/blob/d030c752e5b258f61a225612ce1767e5301ac790/LICENCE"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wouldn't a permalink be preferable though? Was trying to avoid link rot, that was what I got when I clicked "permalink" on the file in github

@AmitMY AmitMY merged commit a433a17 into sign-language-processing:master Jun 20, 2024
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Add LSA-T: "The first continuous Argentinian Sign Language dataset for Sign Language Translation"
2 participants