Monitoring Object Detection Abnormalities via Data-Label and Post-Algorithm Abstractions

TL;DR

This paper introduces an abstraction-based framework for monitoring object detection in autonomous vehicles, aiming to identify unreliable detections through data-label and post-algorithm abstractions, validated on public datasets.

Contribution

It proposes a novel logical framework using data-label and post-algorithm abstractions to filter erroneous object detection results in autonomous vehicle systems.

Findings

Effective filtering of unreliable detections demonstrated

Framework validated on publicly available datasets

Improves robustness of object detection modules

Abstract

While object detection modules are essential functionalities for any autonomous vehicle, the performance of such modules that are implemented using deep neural networks can be, in many cases, unreliable. In this paper, we develop abstraction-based monitoring as a logical framework for filtering potentially erroneous detection results. Concretely, we consider two types of abstraction, namely data-label abstraction and post-algorithm abstraction. Operated on the training dataset, the construction of data-label abstraction iterates each input, aggregates region-wise information over its associated labels, and stores the vector under a finite history length. Post-algorithm abstraction builds an abstract transformer for the tracking algorithm. Elements being associated together by the abstract transformer can be checked against consistency over their original values. We have implemented the overall framework to a research prototype and validated it using publicly available object detection datasets.