Image Classification on the Edge for Fast Multi-Camera Object Tracking

This paper introduces a stochastic model for testing a low-latency method of tracking an object as it moves through an area observed by a dense network of video surveillance cameras. This new method utilizes the computing power of edge devices closer to the source of the video data to run lightweight image classifiers. The sensor redundancy in wide camera networks allows us to increase the accuracy of lightweight image classifiers and provide a real-time estimate of a target's location in the sensing region. Running image classifiers on the edge eliminates the latency costs of ofloading all video data frames to the cloud.