What should be ML(Machine Learning) learning path to extract foreground from video?

Hi there. I am new for Machine Learning. I am trying to figure out which part of the machine learning should I learn to extract foreground of the video. I have tried extract video foreground with image processing libraries such as OpenCV.I have used the following algorithms:

  • Gaussian Mixture-based Background/Foreground Segmentation
  • Grabcut

But results are not efficient. So I want to give a try with Artificial Intelligence. As a beginner in Machine Learning I don’t know what to learn and what not to learn to achieve ability to build this project, what should I learn to implement Machine Learning to extract foreground from video?

Hi there! I don’t have any experience in video image extraction, but this paper (PDF) may at least be a starting point.

Good luck!