As a group of international students in Hong Kong, it is difficult to explain our choice of hairstyle. This application was aiming to make it easy and that is where the inspiration comes from, to make this application.

What was the aim

The android application would allow the user to try different hairstyles to find the one that suits your face by actually placing the hairstyle in real time.

How does it work

This is possible using the Face++ API that helps us recognise different landmarks on the face and perfectly fit the hairstyle you choose, on your face.

Main challenges in this journey

The main challenge of this project was to successfully send requests to the Face++ API and make an android app to communicate with the back-end server that is written in Node.js.

The feeling of accomplishment

It is a sense of accomplishment to have managed to accurately processed images through Face++ API. This showed that it was indeed possible for us to make such an application.

Our learning outcomes

In this journey, I have learnt a lot about image processing. I also managed to apply a lot of knowledge that I had learned in introductory classes in programming and computer vision.

Plans for the Future

We plan to complete the android application and attempt to make an iOS application too. While we have still not managed to create a complete android application, we would continue this project in the future to do so.

How to use the demo

The demo is a URL of my application that returns the coordinates on the image which will be helpful data for placing our hairstyle templates on the user's face, which is near the ears. The request must be a POST request to the server with the following JSON object:

    'img' : '<image in base_64 string format>'

The response of this will be another JSON object with x,y coordinates of the two points:

                'p1x': xleft,
                'p1y': yleft,
                'p2x': xright,
                'p2y': yright

Built With

Share this project: