Surprisingly a lot of time was spent on thinking of an app to create. This took much more time then expected originally We were interested in using facial recognition to build our app. We reflected on our daily life, and realized that we have to search for specific songs depending on the mood and this could be time consuming as you have to look through so many songs. So we decided to built an app that takes a quick picture of you and suggests a song based on your mood!
What it does
M00dReader requires the user to take a picture with a camera, which then gets analysed by an API from which the returned data is coded to assign the facial features of a person to a specific mood. The the mood is used to look for a song and recommends it to the user.
How we built it
Challenges we ran into
A lot of time was spent thinking about an idea to create, since most of our ideas were already created by someone on the web. Finally once we decided to use our idea of using facial recognition, we had troubles looking up for an appropriate API that does what we needed it to. We had then were running out of online storage for pictures that were being taken by the camera during the troubleshooting process. After that, a lot of time was spent on troubleshooting on why data was not parsing properly to be read and output after manipulating it.
Accomplishments that we're proud of
There are many accomplishments we were proud of throughout our project. The main one being using the API and inputting the picture into it and getting the required output from it. This took most of our time during the building as we had to start with inputting a facial picture into the API, then using JSON to make out data readable and then using JSON again to get an output.
What we learned
What's next for M00dReader
M00dReader does have the potential to be large-scale as we plan to implement more moods into the app. We would implement to use machine learning to detect human emotions better and result in more accurate results. Also, we would add a larger dynamic playlist for songs that adds on to itself depending on current popular songs.