Difference between revisions of "Artificial Intelligence"

From Open Food Facts wiki
Jump to navigation Jump to search
 
(14 intermediate revisions by 2 users not shown)
Line 1: Line 1:
Machine learning models and "AI" are used in Open Food Facts to extract information from product images and to perform predictions, such as the product category.
+
[[Category:Artificial Intelligence]]
 
+
= 📺 Presentation =
 +
* Watch Raphael explain (in French) how we use Artificial Intelligence at Open Food Facts: https://tube.numerique.gouv.fr/w/dEHLAbuchWu8zGzY5JgSgz
 
= Principles =
 
= Principles =
 
* AI is there to help accelerate our transparency mission by augmenting Open Food Facts contributors, letting them focus on value-added tasks and increasing their personal impact.
 
* AI is there to help accelerate our transparency mission by augmenting Open Food Facts contributors, letting them focus on value-added tasks and increasing their personal impact.
Line 6: Line 7:
 
* Even if a model does not have perfect results, at Open Food Facts scale, it might save contribution time and increase food transparency. Furthermore, it's possible to leverage the community to validate guesses and improve the models over time.
 
* Even if a model does not have perfect results, at Open Food Facts scale, it might save contribution time and increase food transparency. Furthermore, it's possible to leverage the community to validate guesses and improve the models over time.
  
 +
== Key dates and numbers ==
 +
== Why it matters ==
 +
* Machine learning models and "AI" are used in Open Food Facts to extract information from product images and to perform predictions, such as the product category.
 +
* AI is there to help accelerate our transparency mission by augmenting Open Food Facts contributors, letting them focus on value-added tasks and increasing their personal impact.
 +
== What we’ve done recently ==
 +
* https://github.com/openfoodfacts/robotoff/releases
 +
* https://github.com/openfoodfacts/openfoodfacts-ai/releases
 +
* https://huggingface.co/openfoodfacts
  
 +
== What we are currently working on ==
 
= Projects =
 
= Projects =
  
== Robotoff ==  
+
== 🤖 Robotoff ==  
 +
 
 +
[https://github.com/openfoodfacts/robotoff Robotoff] is Open Food Facts "AI" service: it takes care of calling the ML models and storing all predictions. [https://openfoodfacts.github.io/robotoff/ Robotoff documentation] describes in details how Robotoff works.
  
[https://github.com/openfoodfacts/robotoff Robotoff] is Open Food Facts "AI" service: it takes care of calling the ML models and storing all predictions. [https://github.com/openfoodfacts/robotoff/wiki Robotoff wiki] describes in details how Robotoff works.
 
 
Depending on its confidence, each prediction is either applied automatically or require a manual validation. [https://hunger.openfoodfacts.org/ Hunger Games] is our in-house validation tool.
 
Depending on its confidence, each prediction is either applied automatically or require a manual validation. [https://hunger.openfoodfacts.org/ Hunger Games] is our in-house validation tool.
  
 
All machine learning models used in production are stored on Github as releases, in the [https://github.com/openfoodfacts/robotoff-models/releases robotoff-models] repository.
 
All machine learning models used in production are stored on Github as releases, in the [https://github.com/openfoodfacts/robotoff-models/releases robotoff-models] repository.
 +
 +
Issues opened in [https://github.com/openfoodfacts/robotoff/issues Robotoff repository] are only related to Robotoff codebase improvements and bug tracking. To discuss machine learning projects or tasks, use the [https://github.com/openfoodfacts/openfoodfacts-ai/issues openfoodfacts-ai issue tracker]. We also store research code in the openfoodfacts-ai repository.
  
 
== Hunger Games ==
 
== Hunger Games ==
 
Hunger Games is our collaborative and playful annotation engine. It allows you to annotate in bulk logos and labels, to mass complete nutrition tables and ingredient lists, and to answer Robotoff questions in vast numbers.
 
Hunger Games is our collaborative and playful annotation engine. It allows you to annotate in bulk logos and labels, to mass complete nutrition tables and ingredient lists, and to answer Robotoff questions in vast numbers.
  
== Models ==
+
== Category prediction ==
 +
* More details at: https://openfoodfacts.github.io/robotoff/explanations/category-prediction/
  
A list of our AI projects (current and future) can be found on Github: [https://github.com/openfoodfacts/openfoodfacts-ai/issues openfoodfacts-ai].
+
== Logo and label detection ==
 +
* More details at: https://openfoodfacts.github.io/robotoff/research/logo-detection/ann-benchmark/
  
=== Nutri-Score prediction ===
+
== Nutrition table extraction ==
Our classic model that's able to predict the letter of the Nutri-Score on the packaging is soon to be obsoleted by our more generic logo detection model.
+
* Ongoing project at https://wiki.openfoodfacts.org/NutriSight
  
=== Category prediction ===
+
== Ingredient autoselection ==
 +
* TODO: Doc me
  
=== Logos and labels detection ===
+
== Spellcheck project ==
 +
* [[✍️ Ingredients Spellcheck]]
  
=== Nutrition facts table data extraction ===
+
== What’s coming up next ==
 
+
* We have a lot of potential impactful projects. Please get in touch.
The [[Nutrition facts table data extraction]] project aims to automatically extract the nutrition facts values from photos of nutrition facts table.
+
== How to get involved ==
 
+
== How to contribute? ==
=== Ingredients spellchecking and correction ===
 
 
 
= How to contribute? =
 
  
 
There are many ways to help:
 
There are many ways to help:
Line 43: Line 57:
  
 
Most of our discussions happen on Slack.
 
Most of our discussions happen on Slack.
Depending on your interests, you can join the following channels:
+
Depending on your interests, you can join the following channels: #robotoff and #hunger-games (robotoff and hunger games channels respectively).
* #robotoff and #hunger-games (robotoff and hunger games channels respectively)
+
 
* #ai-machinelearning (generic machine learning discussions)
+
 
* #ai-logo-detection, #ai-imagerecognition, #ai-computervision, #spellcheck
+
== 📆 Weekly meeting ==
 +
* You're also welcome to join the Robotoff weekly e-meeting that takes place every Tuesday at 11am CET (Paris time): https://meet.google.com/qvv-grzm-gzb
 +
* For convenience, you can subscribe to the Community Calendar: [[Events]] to get a weekly reminder
 +
== Get in touch ==
 +
{{Box
 +
| 1    =  Slack channel
 +
| 2    =  [https://openfoodfacts.slack.com/messages/CF0286BRQ/ #robotoff]
 +
}}

Latest revision as of 10:45, 6 September 2024

📺 Presentation

Principles

  • AI is there to help accelerate our transparency mission by augmenting Open Food Facts contributors, letting them focus on value-added tasks and increasing their personal impact.
  • We can use AI and non-AI techniques together and blend them, so that when one of the technique fails for some reason, the other one can save the day.
  • Even if a model does not have perfect results, at Open Food Facts scale, it might save contribution time and increase food transparency. Furthermore, it's possible to leverage the community to validate guesses and improve the models over time.

Key dates and numbers

Why it matters

  • Machine learning models and "AI" are used in Open Food Facts to extract information from product images and to perform predictions, such as the product category.
  • AI is there to help accelerate our transparency mission by augmenting Open Food Facts contributors, letting them focus on value-added tasks and increasing their personal impact.

What we’ve done recently

What we are currently working on

Projects

🤖 Robotoff

Robotoff is Open Food Facts "AI" service: it takes care of calling the ML models and storing all predictions. Robotoff documentation describes in details how Robotoff works.

Depending on its confidence, each prediction is either applied automatically or require a manual validation. Hunger Games is our in-house validation tool.

All machine learning models used in production are stored on Github as releases, in the robotoff-models repository.

Issues opened in Robotoff repository are only related to Robotoff codebase improvements and bug tracking. To discuss machine learning projects or tasks, use the openfoodfacts-ai issue tracker. We also store research code in the openfoodfacts-ai repository.

Hunger Games

Hunger Games is our collaborative and playful annotation engine. It allows you to annotate in bulk logos and labels, to mass complete nutrition tables and ingredient lists, and to answer Robotoff questions in vast numbers.

Category prediction

Logo and label detection

Nutrition table extraction

Ingredient autoselection

  • TODO: Doc me

Spellcheck project

What’s coming up next

  • We have a lot of potential impactful projects. Please get in touch.

How to get involved

How to contribute?

There are many ways to help:

Most of our discussions happen on Slack. Depending on your interests, you can join the following channels: #robotoff and #hunger-games (robotoff and hunger games channels respectively).


📆 Weekly meeting

  • You're also welcome to join the Robotoff weekly e-meeting that takes place every Tuesday at 11am CET (Paris time): https://meet.google.com/qvv-grzm-gzb
  • For convenience, you can subscribe to the Community Calendar: Events to get a weekly reminder

Get in touch

Slack channel