Upload a picture of your palm and find out whose lifepath will you follow...
What it does
It takes a picture of your palm and predicts which famous people you are similar to. Based on that you find out some funny facts that are going to happen to you.
How we built it
We built a mobile application using Cordova (PhoneGap) framework. Clients using mobile devices communicate with that is server implemented in Python library Flask and is deployed on Heroku by sending images to server and receiving predictions about their future from it. Server processes the images with the help of OpenCV image manipulation library with the intention of feature extraction. Based on a model which was built with KNN (k-Nearest Neighbors) machine learning algorithm beforehand, we predicted which persons' palm the processed image is most similar to. The information about three top predictions is sent back to the client and presented to the user. The model itself was based on the dataset we prepared ourselves by hand. We downloaded a lot of images of famous people's palms, transformed them into an appropriate shape / form and used them as our training set.
Challenges we ran into
We faced a lot of difficulties setting up Python environment on Heroku. There were many obstacles we had to overcome when transferring the image from the client to the server.
Accomplishments that we're proud of
We built a working app in just 24 hours and surprised ourselves.
What we learned
We learned a lot about deploying applications to Heroku.
What's next for Stars of Platypus