Allow to select some photos from the gallery app and "Share them to Open Food Facts"

openfoodfacts / smooth-app

🤳🥫 The new Open Food Facts mobile application for Android and iOS, crafted with Flutter and Dart

https://world.openfoodfacts.org/open-food-facts-mobile-app?utm_source=off&utf_medium=web&utm_campaign=github-repo

Apache License 2.0

844 stars 282 forks source link

Allow to select some photos from the gallery app and "Share them to Open Food Facts" #4334

Open teolemon opened 1 year ago

teolemon commented 1 year ago

Problem

You might have product photos in your camera roll for various reasons.
It might be too long to associate them manually with a barcode using the current picker.
Allow to select some photos from the gallery app and "Share them to Open Food Facts", where a barcode detection will be done to propose a product to upload them to.
Implementation inspiration can be drawn from the old app

monsieurtanuki commented 1 year ago

@teolemon Very interesting UX I guess! Btw it looks like you're in Sibiu, which is a really lovely place.

teolemon commented 1 month ago

Google just launched "Share to Gemini" to share any image to Gemini, and ask it a question (eg: "is it vegan") https://www.androidpolice.com/searching-images-gemini-android-just-got-better/

monsieurtanuki commented 1 month ago

Google just launched "Share to Gemini" to share any image to Gemini, and ask it a question (eg: "is it vegan") https://www.androidpolice.com/searching-images-gemini-android-just-got-better/

Of course that works only if you have the Gemini app installed. Not like an API call, I mean.

teolemon commented 1 month ago

Yes, it was more the acquisition channel I was talking about.

In the old app, we used Share to Open Food Facts for product addition.
We could reintroduce that as proposed above, but we could potentially open the product if we detect a barcode, or extract /parse ingredients if we detect ingredients (we could have a on-device version of our recently introduced image classifier)
That's currently not a priority, but that would provide more or less the same service, at a lower computational cost.