My Android app takes a picture and tells the user what is in the picture.
- User takes a picture of a dog driving a car near the Eiffel Tower.
- My app says: "dog; car; eiffel tower".
- Must not rely on a web service, because the app must be able to work without an Internet connection.
- No need to "train" it, the library already includes the results of the training.
- 100% open source (training data and training software do not have to be open source, but the final software including data resulting from the training must be open source)
- The objects to recognize are objects frequently seen by humans (dog), types of landscapes (forest), famous places (Eiffel Tower).
- Super-bonus if the keywords are Wikidata objects (dog → https://www.wikidata.org/wiki/Q144)
- Bonus for providing a probability with each keyword (97% chances that this picture shows a dog)
- Preferably the library should not be bigger than 3 MB, to not make the app a pain to download.