It is no secret that the watch industry is doing better and better with a watch that becomes a fashion effect, a style effect, a taste effect, an investment practice while keeping its original practicality: knowing the time at any time of the day

Although the Swiss watch industry, especially from luxury watches, is largely hegemonic with more than 53% of the international market with only 2% of the world volume produced, we offer a completely new application for you, the watch fans.

Our idea is to offer customers the opportunity to download an application that allows them to find out about the specifics of the object they want to find out about, in this case the watch they like.

The concept is rather simple: once the application is downloaded, the customer only needs to scan the watch (on which he wishes to find out more) and the application then provides him a business card of this watch with: the characteristics (diameters, movement, type, bracelet materials, type of dial, water resistance, ...) as well as the prestige of the watch (limited edition?), the price at the supplier's and the general rating of the watch in question (out of 10). Then you can either buy it, save it for later or just find out more about it.

This application is not aimed at a specific target audience, as anyone can find out about watches, although it will undoubtedly be used more by watch enthusiasts and fans.

However, we can ask ourselves the WHY of such an application. You have probably already seen a watch you like on a friend, on a passer-by, on a salesman, in a shop window but either because of lack of time, shyness or forgetfulness, you do not get the necessary information about this watch. This is where the application comes into play: thanks to a 10-second manipulation, you get all the necessary information about the watch in order to make your choice as easy as possible. The aim of this application is therefore to be as close as possible to the consumer.

So, are you ready to use this app?

You are no doubt familiar with the famous Shazam and Vivino applications: Shazam is an "Apple-owned mobile application used to identify songs based on a small sample using the microphone inside a user's device" and Vivino which is a wine app that "scans wine labels and provides details about that wine such as user and expert ratings, price, features, and the nearest store where you can buy it or purchase and order directly on your phone".

We can possibly describe these two applications as direct competitors of our application because although they do not offer the same service, for a different product range, the technology used to operate these applications bears a strong resemblance to our own, especially for Vivino. If Shazam and Vivino were to start selling their technology and the way they operate then we would have even more competitors.

We can therefore wonder about the technology used in our application. To begin with, thanks to machine learning, it is now possible to create such an application. Machine learning corresponds to the study of computer algorithms that are perfected on their own through experience. It is considered as part of artificial intelligence since they build a model which itself builds sample data in order to make predictions, to take decisions. Machine learning algorithms are used in many applications (e-mail filtering, artificial vision for example). In particular to use this, it is necessary to use ml5.js which aims to make machine learning accessible to a large audience of students, artists, ...

For example, I will show you how and by what means the AI manages to recognise images (either by loading them from our computer or from the camera) and give the percentage of matches.

To do this, you need to configure ml5.js and MobileNet to classify an image, so I'm going to present you with a code that can help you get started:

Taken From

You have well understood that for watches we are going to use the same procedure: the piece of code will allow us to recognise the object as a watch because we will have taken the photo of the watch thanks to our telephone. The complexification of the code will then make it possible to distinguish the different watches according to their characteristics. So, if you would like to learn more about the subject, here are a few links that may help you in your research and approach: