ComPrivDet: Efficient Privacy Object Detection in Compressed Domains Through Inference Reuse

TL;DR

ComPrivDet is an efficient method for detecting privacy objects in compressed videos by reusing inference results, significantly reducing latency while maintaining high accuracy.

Contribution

It introduces a novel approach to privacy object detection in compressed videos that reduces inference overhead by reusing previous results and selectively skipping frames.

Findings

Maintains 99.75% accuracy for face detection

Achieves 96.83% accuracy for license plate detection

Reduces inference latency by 75.95% compared to existing methods

Abstract

As the Internet of Things (IoT) becomes deeply embedded in daily life, users are increasingly concerned about privacy leakage, especially from video data. Since frame-by-frame protection in large-scale video analytics (e.g., smart communities) introduces significant latency, a more efficient solution is to selectively protect frames containing privacy objects (e.g., faces). Existing object detectors require fully decoded videos or per-frame processing in compressed videos, leading to decoding overhead or reduced accuracy. Therefore, we propose ComPrivDet, an efficient method for detecting privacy objects in compressed video by reusing I-frame inference results. By identifying the presence of new objects through compressed-domain cues, ComPrivDet either skips P- and B-frame detections or efficiently refines them with a lightweight detector. ComPrivDet maintains 99.75% accuracy in private face detection and 96.83% in private license plate detection while skipping over 80% of inferences. It averages 9.84% higher accuracy with 75.95% lower latency than existing compressed-domain detection methods.