EmoSuggest

Inspiration

We were inspired by the growing interest towards text and image recognition and the fact that this technology could be used as a tool for improvement in human experience.

What it does

EmoSuggest gives you an option to either take a picture or choose from one of your existing ones and detect the emotion from it so that your hands could take a rest for a while. Based on the emotion [anger, sadness, happiness, disgust, contempt, fear, surprise] detected, it suggests the user Music, Videos and News as of now that could help soothe the user's mood.

How we built it

We built using the Android Studio taking help of multiple open source APIs from Youtube Data, Spotify, Google News, and Washington Post. On the integration of these APIs into the face detection API we were successfully able to detect emotions from a picture and then link the responses of these APIs in a meaningful way.

Challenges we ran into

Being new to Android Development, there was a deep learning curve for the team. Integrating various SDKs and REST APIs into the Android application was tricky because of the authentication needed by all the third party services and integrating those client keys into the application was time taking.

Accomplishments that we're proud of

We were successfully able to completely build the application, test it for various image datasets and design the UI in an intuitive manner. Building something from scratch while not being aware of the nuances of Android Studio and helping each other throughout the short period of time was definitely worth remembering and something to be proud of.

What we learned

Technically, the team learned to work with various third party services, Android Studio, Authentication mechanisms, UI design, FireBase, Azure.

What's next for EmoSuggest

Currently, EmoSuggest gives out only Music, Video and News suggestions. The idea is to extend this to other areas, and also to give a number of options to choose from. Integration with Ok Google and Amazon Alexa are also places to be explored into. Finally, let the user have some amount of control on what kind of suggestions he would be more relatable to and try and give him a better experience.