Inspiration
This project was inspired by a snapchat filter, which tells you how attractive you are. So I did some research What makes a face beautiful?, So I programmed my on "filter" which tells you how attractive you are in a percentage by detecting how symmetrical the faces are.
What it does
This python project, will use opencv and mediapipe to detect how attractive the detected faces are, by detecting how symmetrical the detected faces are.
How we built it
We used the opencv module and the mediapipe module. The MediaPipe module is a machine learning module developed by Google that provides a collection of functions for creating a wide range of computer vision applications.
Challenges we ran into
- Lighting Condition: The lighting conditions can affect the accuracy of the face detection algorithm. If the lighting is too bright or too dark, it can cause the algorithm to miss some faces or detect false positives.
- Face orientation: The face detection algorithm may not work well if the face is not oriented in a certain way. For example, if the face is tilted too much or if it’s not facing the camera directly, it can cause the algorithm to miss some features.
Accomplishments that we're proud of
I can be proud of several accomplishments after writing a program that uses OpenCV and MediaPipe to detect how symmetrical the detected faces are on the live camera feed. For example, I learned new skills while writing the program, such as how to use OpenCV and MediaPipe. Finally, I improved my programming skills in general by learning how to use these powerful tools for computer vision and machine learning.
What we learned
I became more familliar with theese modules.
What's next for Face Symmetry Detector - Python
This project, is just one of the many computer vision opencv python projects that I have made.
Log in or sign up for Devpost to join the conversation.