Skip to content

ankitgoyal0301/COVID-19-Face-Mask-Detector

Repository files navigation

COVID-19-Face-Mask-Detector

About the Project:

  • Realtime face mask detector using Keras, Deep Learning & OpenCV.
  • Used opencv’s Object detection techniques to detect faces.
  • Transfer Learning using MobileNetV2 architecture.
  • Fine-tuning the model using Data Augmentation.

Training and Validation Plots:

         Accuracy:                 Loss:

Accuracy Loss

Output of the model over images:

Sample Input 1     Sample Output 1

Sample Input 1     Sample Output 1

Sample Input 1     Sample Output 1

Sample Input 1     Sample Output 1

Output of the model over Video file:

Input  Output

Refer the full video inputs and outputs here:

Input Video

Output Video

Limitations of the current project:

  • Facing difficulties in detecting faces with low light and low quality webcams.
  • Dataset has images with people wearing only white colored masks 😷
  • Face detection can be improved.

About

COVID-19 Face Mask Detection system based on computer vision and deep learning using OpenCV and Tensorflow/Keras.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published