Edge-Assisted On-Device Model Update for Video Analytics in Adverse Environments

TL;DR

This paper introduces an edge-assisted framework for continuously updating lightweight video analytics models on resource-constrained devices, significantly improving accuracy in adverse environments while reducing retraining time.

Contribution

The proposed system uniquely combines key frame extraction, trigger-based retraining, and edge server updates to enhance on-device video analytics in challenging conditions.

Findings

Improves accuracy by up to 24% in adverse environments

Reduces retraining time by over 50% compared to benchmarks

Demonstrates effectiveness on devices with different capacities

Abstract

While large deep neural networks excel at general video analytics tasks, the significant demand on computing capacity makes them infeasible for real-time inference on resource-constrained end cam-eras. In this paper, we propose an edge-assisted framework that continuously updates the lightweight model deployed on the end cameras to achieve accurate predictions in adverse environments. This framework consists of three modules, namely, a key frame extractor, a trigger controller, and a retraining manager. The low-cost key frame extractor obtains frames that can best represent the current environment. Those frames are then transmitted and buffered as the retraining data for model update at the edge server. Once the trigger controller detects a significant accuracy drop in the selected frames, the retraining manager outputs the optimal retraining configuration balancing the accuracy and time cost. We prototype our system on two end devices of different computing capacities with one edge server. The results demonstrate that our approach significantly improves accuracy across all tested adverse environment scenarios (up to 24%) and reduces more than 50% of the retraining time compared to existing benchmarks.