Find, Go and Taste

The idea

In the course of designing the web, I choose the module of Artificial Intelligence, which is a mainstream in the future.

The aim of the techonology is to improve people's life quality. I wonder how this techonology can bring different experience to us. Then I find that everyone needs to eat --- that means, we need and enjoy the food, which may brings us pleasure everyday.

Target

To find the essence of the food, foodies try to discover in different ways, videos, photos, documentary films etc. However, when foodies get interested in a kind of food, which is a photo, they don't know the name and the stories behind the food. What a pity it is!

To satisfy the foodies' curiosity and encourage them to travel and taste the food around the world, I decide to develop an application which can help the foodie find the names, stories and the original place of certain kind of food. When they upload the picture of a certain food, they may have a deeper understanding and even gain a new experience.

Why choose our application

What? How?

Maybe you consider that it is easy to find the information about food through searching engine. However......

Imaging that your friends share some food photos they try when travelling, but they don't introduce and you want to know more about the food...

Imaging that you find a special food in a film or video, but there is no description...

How to code

// Step 1: Using MobileNet to create an image classification model
const classifier = ml5.imageClassifier('MobileNet', onModelReady);

// Step 2: Choose an image
const img = document.querySelector("#myImage")

// Step 3: Predicting image classification results
let prediction = classifier.predict(img, gotResults);

// Step 4: Manipulating results
function gotResults(err, results) { console.log(results); }

Resources and Reference

Books, films, articles...

The basic knowledge of ml5.js
Friendly Open Source Machine Learning Library for the Web
Building an insanely fast image classifier with MobileNets
Creating insanely fast image classifiers with MobileNet in TensorFlow

Let's do this!