- Import and read the video, extract frames from it, and save them as images.
- Label a few images for training the model.
- Build our model on training data.
- Make predictions for the remaining images.
- Calculate the screen time of the actors.
- Detecting a face using haarcascade
- Forming a square around any face detected.
- Enter id for every different face and gather images of faces.
- Train on the basis of gathered data
- Here I am using Local Binary Pattern Histograms Algorithm
- Providing a different video section and recognizing the actor by name