Tracking and Blurring the Face in a Video File

Abstract

This paper shows how to detect, track and blur a face in video frame using the Viola-Jones detection algorithm for detection and KLT algorithm to tracks a set of feature points across the video frames. First, the detection locates the face, and then identifies feature points that can be reliably tracked.

A blurring filter is applied after detection and tracking the face and very good results are obtained that can be useful to blurring the face; for example, when showing a painful view that includes distorted faces in wars reports or crimes.