Inspiration
Joaquin Phoenix, aka The Joker was our inspiration. Several of the members of our group have extreme appreciation for the arts and we were all greatly amazed by the portrayal of sheer madness in The Joker.
What it does
Our project takes a face that exists in pure sanity and distorts the image, live or static, into madness. It uses OpenCV for image processing and a pre-trained machine learning model for Computer Vision facial detection capabilities.
How we built it
We built it using TypeScript to create an interactive user interface and we used Python, specifically through a Flask web server for the backend, leveraging several libraries.
Challenges we ran into
We ran into several challenges getting the project to send live video feedback over the Flask api. We also ran into challenges trying to get the project to properly capture and distinguish between emotions that were signified by slightly similar facial expressions.
Accomplishments that we're proud of
We are very proud of the fact that we were able to produce a bug-free, production ready project that is robust.
What we learned
We learned that it is extremely important to have visibility into the responses sent and received by the backend server and to include as much error logging as possible.
What's next for Horror-Vision
What's next for Horror Vision is the transformation into a full-fledged method acting tool that hollywood actors can utilize to train for roles such as horror movies and they can be graded on their ability to display specific emotions through varying facial expressions.
Built With
- computervision
- flask
- github
- opencv
- python
- render
- typescript
Log in or sign up for Devpost to join the conversation.