Inspiration
All members in our team like fashion and most of the time we want to find similar items of what we like for later on purchase them.
What it does
Users upload a photo of their favorite clothes and Clothify find similar items for the user to choose. Clothify extract all the relevant information of the products in order to be purchased with ease.
How I built it
We developed an android app as the client and a robust API developed with python following the serverless architecture, host on AWS and secured by identity and user pools.
For achieving the suggestions we work with Image Classification using Convolutional Neural Networks — Deep Learning in python.
And for gathering the data we used beautiful soup 4 to scrape web sites.
Challenges I ran into
We faced a lot of challenges: Creating the data sets, training the networks and scrapping websites.
Accomplishments that I'm proud of
We a 100% functional api and application. We where able to create an intelligent service for classifying fashion.
What I learned
A lot python data science.
What's next for clothify
Make it more precise in order to upload it to the stores.
Built With
- amazon-dynamodb
- amazon-web-services
- android
- beautiful-soup
- java
- keras
- python
- serverless-framework
- tensorflow
Log in or sign up for Devpost to join the conversation.