How we built it
We used html to build the prototype and python for the face/skin/skin tone detection framework with convolution neural networks to classify features in images and generate personalized seasons and color palettes.
Challenges we ran into
Learning new languages: python and html Figuring out how to execute our concept. Finding and training data. Figuring out how to make frameworks for skin detection through webcams in real time.
Accomplishments that we're proud of
Learning python and html. Working together to complete a project in 24 hours.
What we learned
Learned how to code a full project. Having an idea and executing it are two completely different things.
What's next for Seasons by Aritzia
This product can further be improved to include undertones and hair colour to provide users with more accurate results as well as maintaining this product’s quality is vital to its effectiveness.
We acknowledge that there are some limitations to this product, including the racial bias that is present in databases which translates to the machine learning system. This can be mitigated by continuing to improve and include a wider range of datasets for the algorithm.
Log in or sign up for Devpost to join the conversation.