That's the idea behind TheMirror, a personal project I'm building to help customers visualise hairstyles before making a decision.

I'm documenting my journey from the first lines of Python to building a working application, including what I'm learning, the challenges I face, and the technical decisions I make along the way.

My ambition is to become an AI engineer, and I believe the best way to learn is by building things that solve real-world problems.

Here's a look at Week 1: laying the foundations.

The Problem: Preview Before Making a Decision

At a hair salon, a customer might arrive with a picture of a hairstyle they like. They might also choose a style from the catalogue offered by the hairstylist.

However, there is a significant difference between seeing a hairstyle in a picture and imagining what it would look like on yourself.

This is the gap I want to address with TheMirror.

The idea is to allow a customer to select a hairstyle and then preview it using their own image. They could then share their opinion on the result before making a decision together with the hairstylist.

The goal is not to replace professional advice or make decisions on behalf of the customer.

TheMirror is simply intended to become a visualisation tool that encourages communication between the customer and the hairstylist.

The project is publicly available on my GitHub repository:

🔗 https://github.com/jinonoa/TheMirror

My Vision for TheMirror

The functionality I currently have in mind is based on several elements:

• A hairstyle catalogue that allows styles to be added, edited or deleted;
• The customer selecting a hairstyle;
• Taking a picture of the customer's head;
• Face detection;
• Previewing the selected hairstyle;
• Giving the customer the opportunity to share their opinion.

Ultimately, I want to create an experience that is simple enough for customers to explore different options without needing any technical knowledge.

An excerpt from my README already captures part of this vision:

Functionality

• Adding, editing or deleting a template from the catalogue
• Model selected by the customer
• Pan shot of the customer's head

The project is still in its early stages, but this initial description gives me a starting point from which to gradually build the application.

From an Idea to a Visual Concept

Before diving deeper into the technical side of TheMirror, I wanted to visualise what the final application could look like.

These initial sketches represent my vision for the user experience, from selecting a hairstyle to previewing it and sharing feedback.

They are not final designs, but rather a starting point to help me organise my ideas and think about how customers will interact with the application.

My goal is to make the experience as intuitive as possible, allowing customers to explore different hairstyles without unnecessary complexity.

Here is an early look at what I have in mind for TheMirror.

A Project Connected to My Ambitions in AI

I want to become an artificial intelligence engineer, with a particular ambition to work on AI applied to video games and interactive experiences.

TheMirror allows me to explore another area of AI while developing a skill I consider essential: starting with a real-world problem and figuring out how to turn it into a technical solution.

Building this application will require me to gradually understand different technologies related to computer vision and image processing.

In particular, I need to learn how to:

• Detect a face in an image;
• Identify useful information from an image;
• Work with images and AI models;
• Understand which technologies are suitable for my problem;
• Integrate these technologies into a real application.

My goal is not simply to use an existing model. I want to understand what I am doing well enough to build a coherent solution around it.

Week 1: Laying the Foundations

During the first week of working on TheMirror, I mainly focused on laying the project's foundations and strengthening my Python skills.

I started by installing and testing my development tools: VS Code, Python, Git and Docker. I then created the public TheMirror repository on GitHub and began writing its README to explain the project's purpose and objectives.

To better understand the needs the application would have to address, I also spoke with people who could provide insights into the hairstyles and colours most frequently requested by customers. These conversations helped me put together an initial catalogue of ten reference photos.

Alongside this, I completed several exercises on Exercism and Jiki to strengthen my Python fundamentals. I worked on variables, functions, Boolean operators, numbers, calculations, function calls and conditional statements.

These exercises presented several challenges involving understanding and logic, particularly when designing functions, using conditions and applying certain mathematical formulas. I used documentation and AI tools to understand my mistakes rather than simply copying ready-made solutions.

Finally, I began researching tools for face detection, image analysis and initial computer vision experiments. I also started thinking about the application's overall architecture and user interface.

This first week allowed me to establish TheMirror's technical and functional foundations while identifying the skills I will need to develop further to turn the idea into a working prototype.

Here are a few screenshots from my initial experiments and development environment:

What I Still Need to Learn

TheMirror is still a long way from being finished, and that is precisely what makes this project interesting.

I do not yet know:

• Which technology to use for face detection;
• How to position a hairstyle correctly on an image;
• How to achieve a sufficiently realistic result;
• How to handle different head orientations;
• How to integrate an AI model efficiently into the application;
• Or which architecture will ultimately be best suited to the project.

I will therefore need to experiment, test different approaches and probably abandon some ideas along the way.

And that is precisely what I am looking for with this project: learning by building something tangible.

What's Next?

After this first week, which was dedicated to laying the foundations, the next step is to gradually move from planning to building a prototype.

I will start by identifying the technologies best suited to face detection and image processing, then experiment with existing solutions before looking for ways to integrate them into TheMirror.

I want to document this progress week by week, sharing the technical decisions, mistakes, experiments and results along the way.

TheMirror starts with a simple problem: allowing a customer to preview a hairstyle before making a decision.

My goal now is to turn this idea into a working first version.

This is just the beginning. More experiments, more challenges, and more progress to come.

If you're interested in AI, computer vision, Python, or building projects from scratch, feel free to follow my journey!