6,323
edits
No edit summary |
No edit summary |
||
Line 1: | Line 1: | ||
== Problem == | |||
* We have a large problem: it's easier to snap pictures than to turn that into text. We thus have a large backlog of products that are not complete. We aim to use clever OCR and techniques, but also Gamification to stay on top. | * We have a large problem: it's easier to snap pictures than to turn that into text. We thus have a large backlog of products that are not complete. We aim to use clever OCR and techniques, but also Gamification to stay on top. | ||
== Best Practises == | |||
* Google Image Labeler and Recaptcha are good examples on how to crowdsource that. | * Google Image Labeler and Recaptcha are good examples on how to crowdsource that. | ||
* https://en.wikipedia.org/wiki/Google_Image_Labeler | ** https://en.wikipedia.org/wiki/Google_Image_Labeler: You basically have players compete to label an image. If they agree on keywords, then the keyword is assumed true and added to the image. | ||
* Wikidata Game |