Utility and Privacy in Object Tracking from Video Stream using Kalman Filter

TL;DR

This paper explores methods to balance privacy and utility in object tracking from video streams using Kalman filters, by controlling localization accuracy to meet privacy constraints.

Contribution

It introduces two novel approaches that regulate localization accuracy to preserve privacy without significantly sacrificing tracking utility.

Findings

Localization accuracy can be effectively bounded using the proposed methods.

The methods ensure privacy constraints are met while maintaining tracking performance.

The approaches are applicable to real-time video stream processing.

Abstract

Tracking objects in Computer Vision is a hard problem. Privacy and utility concerns adds an extra layer of complexity over this problem. In this work we consider the problem of maintaining privacy and utility while tracking an object in a video stream using Kalman filtering. Our first proposed method ensures that the localization accuracy of this object will not improve beyond a certain level. Our second method ensures that the localization accuracy of the same object will always remain under a certain threshold.