Skip to content

Latest commit

 

History

History
38 lines (33 loc) · 2.32 KB

README.md

File metadata and controls

38 lines (33 loc) · 2.32 KB

Perfect-Pause

Computer Vision based attention monitoring system to aid your movie-watching experience.

Link to presentation: https://www.youtube.com/watch?v=FDIZSHDRuGk

An Overview

Our aim is to develop a robust human attention monitoring system based on Computer Vision which will work in harmony with VLC media player, prompting it to execute actions such as pause/play depending on whether the user is paying attention to his monitor or not.

Why is it useful?

  • Our system will ensure that the video/movie keeps playing only when the user is watching.
  • Our system will pause the video/movie when the user is unattentive or is not watching.

Example situations where our model will be useful :

  • If the user falls asleep.
  • If the user has to receive a call/.
  • Attending an emergency chore.
  • Getting called to carry the groceries from the car :)

Our Approach :

We are using openCV’s Haar Cascades determine the ROI (region of interest) and process the inference, all of which takes milliseconds and requires minimal computing power.

What exactly is a Haar-cascade?

  • Haar Cascade is a machine learning object detection algorithm used to identify objects in an image or video and based on the concept of ​​ features proposed by Paul Viola and Michael Jones.
  • It is a machine learning based approach where a cascade function is trained from a lot of positive and negative images. It is then used to detect objects in other images.

Integrating VLC Media player:

We are using a python package called vlc to bridge the gap between Computer-Vision and VLC media player.

A note on privacy :

Our solution does not store or upload any data, it captures frames from the webcam and deletes the frame as soon as it computes the inference.