Towards Building Self-Aware Object Detectors via Reliable Uncertainty Quantification and Calibration

TL;DR

This paper proposes a new framework for testing object detectors' robustness and uncertainty estimation in safety-critical environments, introducing novel metrics, datasets, and a baseline for future research.

Contribution

It introduces the Self-Aware Object Detection (SAOD) framework, addressing deficiencies in current robustness testing and providing a comprehensive benchmark for object detector reliability.

Findings

Object detectors vary significantly in robustness and calibration.

The proposed metrics reveal critical weaknesses in existing detectors.

The baseline improves reliability and calibration in the SAOD framework.

Abstract

The current approach for testing the robustness of object detectors suffers from serious deficiencies such as improper methods of performing out-of-distribution detection and using calibration metrics which do not consider both localisation and classification quality. In this work, we address these issues, and introduce the Self-Aware Object Detection (SAOD) task, a unified testing framework which respects and adheres to the challenges that object detectors face in safety-critical environments such as autonomous driving. Specifically, the SAOD task requires an object detector to be: robust to domain shift; obtain reliable uncertainty estimates for the entire scene; and provide calibrated confidence scores for the detections. We extensively use our framework, which introduces novel metrics and large scale test datasets, to test numerous object detectors in two different use-cases, allowing us to highlight critical insights into their robustness performance. Finally, we introduce a simple baseline for the SAOD task, enabling researchers to benchmark future proposed methods and move towards robust object detectors which are fit for purpose. Code is available at https://github.com/fiveai/saod