What it does: Image classification for commons uploads.
Wiki thing it helps with:
- Crude image categorization (human/selfie, dog, street, house, car) is easier than specific, biggest thing is that you need some train set (thus are not able to predict unseen categories) T331134
- Find uncategorized images
- Find likely unwanted images (copyvio, etc.)
- Estimate image quality T184739
- Estimate image creation date (1920s, 2000s) which could be used to verify PD-old claims
- AI which combines image elements with articles and suggests relevant images (combine text based) T236142
- Automatically generate image captions and alt text
Things that might helps us get this AI built (optional):
- https://commons.wikimedia.org/wiki/User:Basvb/Deeplearning (sort on probability)
- Tensorflow: https://www.tensorflow.org/how_tos/image_retraining/
- https://commons.wikimedia.org/wiki/User:Multichill/Using_OpenCV_to_categorize_files
- https://en.wikipedia.org/wiki/User:DrTrigonBot/doc?rdfrom=commons:User:DrTrigonBot/doc#Categorization
- https://research.googleblog.com/2016/09/show-and-tell-image-captioning-open.html
- Could build a Wikidata-game like interface to provide training data or to assess predictions for inclusion in commons.
- Structured data on commons!
Related efforts:
- T49492: Automatically propose/suggest a category for images
- T76886: Investigate computer vision image classification and description tools for shadow tags and search descriptions
- T135993: catimages: nice to have for the future ("FuDo")
How to move forward?
- Spec out what needs to be done. (We: Commons, WLM, WMF Research and any other community/entity that should be involved).
- WMF Research creates a collaboration to work with one student on the research component. I estimate what we're talking about is a 8-12 week work depending on how much details we want to go into.
- We work with a few developers to implement change. This latter can be a bit more work than usual given that the code-base of Commons will change due to Structured Data on Commons project.