Halloween is approaching and we wanted to make a project that dealt with fear.

What it does

This is a web app that presents a variety of scary images and takes pictures of a user's reaction to those images. Then using Microsoft's Face API, the app calculates the degree of fear generated by the user's face for each image. The app then identifies the most fearful images for that particular person.

How we built it

We used HTML, CSS, and Javascript (jQuery) to build the front-end of our web app. We used Microsoft's Face API to perform sentiment analysis on the images gathered. We also built a prototype Android app that had the same functionality as the web app.

Challenges we ran into

We originally intended to use the Pebble smartwatch to monitor heart rate but later discovered that none of the watches had heart rate sensors. We then tried to measure heart rate using the LED camera of our smartphones, but discovered that implementing this feature was too time consuming and difficult, and there were better ways of measuring distress (using sentiment analysis). We ran into difficulties pushing photos from our smartphone app to the Microsoft Face API.

Accomplishments that we are proud of

We switched from a platform that we were more familiar with (Android development) to a platform that we were less sure about (web development) and still managed to make a working product.

What I learned

We learned how to talk to Microsoft Face API through HTTP. We also learned how to create web apps using HTML/CSS/Javascript.

What's next for Fearface

Implementing cameras from the smartphone or laptops and continuously uploading picture data to the cloud for sentiment analysis. Adding other metrics like heart rate to better track the level of fear. Adding a sort of "streak incentive" so that people would continuously use our app every day. Adding sound so people are more immersed. Maybe a far-fetched idea is to make a VR game (more immersive and intense).

Share this project: