DIALER: A Case for Improving Rare-Class Accuracy in Retraining-Free Edge Video Analytics

2026-10-05

AI Summary

This study addresses the misclassification of rare classes by vision foundation models (VFMs) in edge video analytics, where retraining resources are severely constrained. To this end, we propose a retraining-free dynamic error correction framework. The method constructs a multi-stage correction pipeline offline and introduces a novel approach that leverages idle computational capacity during VFM inference to perform online rare-class rectification and dynamic routing, thereby enhancing accuracy with zero additional overhead. Evaluated on four real-world driving datasets, the proposed framework improves rare-class accuracy by up to 14.0% without compromising real-time inference performance. These results demonstrate that the approach effectively reconciles the competing demands of accuracy and latency in resource-constrained edge scenarios.

Abstract

Edge video analytics with lightweight models is prone to accuracy degradation due to persistent distributional shifts in live video streams. While continuous learning (CL) addresses such data drift, it heavily strains the limited compute resources of edge servers originally provisioned for inference. Our empirical study reveals that emerging vision foundation models (VFMs) offer a practical, retraining-free alternative that delivers high average accuracy with remarkable compute savings. However, VFMs frequently misclassify specific rare classes, which often represent critical objects, as visually similar common classes. We design DIALER, a system that exploits the spare compute cycles freed by retraining-free VFM inference to mitigate rare-class misclassifications. Specifically, DIALER pre-builds multi-stage correction pipelines for dominant rare-to-common confusion pairs offline. At runtime, it routes correction candidates to the corresponding pipelines and executes as many stages as idle GPU headroom permits. Evaluation on four real-world driving datasets shows that DIALER improves rare-class accuracy by up to 14.0% without interfering with real-time VFM inference for multi-stream analytics.

Problem

• Edge Video Analytics
• Vision Foundation Models
• Rare-Class Accuracy
• Misclassification
• Retraining-Free

Innovation

• Edge Video Analytics
• Vision Foundation Models
• Retraining-Free
• Rare-Class Accuracy
• Multi-Stage Correction Pipelines