Recap From Day 040

In day 040, we continued with video features: Haar Cascades. We saw that A Haar Cascade is basically a classifier which is used to detect the object for which it has been trained for, from the source. The Haar Cascade is by superimposing the positive image over a set of negative images. The training is generally done on a server and on various stages. Better results are obtained by using high quality images and increasing the amount of stages for which the classifier is trained.

Today, we will continue from where we left off in day 040

Video Features Continued

Compare pixels from one frame to the next

If we are interested on how things are moving in front of the camera, rather than trying to identify what is in front of a camera, there are a few simple things that might work. One simple thing we can do is compare pixels from one frame to the next.

Source

If someone or something is moving very quickly, we’ll have a lot of change, in RGB or other color space representation of a single pixel from one frame to the next frame. We could some up the magnitude of these changes and get a single feature. Or we could some up these changes for different regions of the scene, and get a few features describing how much motion is happening on the left, in the middle, on the right, and so on.

Awesome. That’s all for day 041. I hope you found this informative. Thank you for taking time out of your schedule and allowing me to be your guide on this journey. And until next time, remain legendary.

Reference

*https://www.kadenze.com/courses/machine-learning-for-musicians-and-artists-v/sessions/sensors-and-features-generating-useful-inputs-for-machine-learning*