Localization Recall Precision (LRP): A New Performance Metric for Object Detection

TL;DR

This paper introduces the Localization Recall Precision (LRP) Error, a new performance metric for object detection that better captures localization accuracy and provides more discriminative evaluation than the traditional Average Precision (AP).

Contribution

The paper proposes LRP Error and Optimal LRP as novel metrics that improve evaluation of object detectors by directly measuring localization and optimizing confidence thresholds.

Findings

Optimal LRP offers richer evaluation information than AP.

Confidence thresholds vary significantly among classes and detectors.

Class-specific thresholds improve detection accuracy.

Abstract

Average precision (AP), the area under the recall-precision (RP) curve, is the standard performance measure for object detection. Despite its wide acceptance, it has a number of shortcomings, the most important of which are (i) the inability to distinguish very different RP curves, and (ii) the lack of directly measuring bounding box localization accuracy. In this paper, we propose 'Localization Recall Precision (LRP) Error', a new metric which we specifically designed for object detection. LRP Error is composed of three components related to localization, false negative (FN) rate and false positive (FP) rate. Based on LRP, we introduce the 'Optimal LRP', the minimum achievable LRP error representing the best achievable configuration of the detector in terms of recall-precision and the tightness of the boxes. In contrast to AP, which considers precisions over the entire recall domain, Optimal LRP determines the 'best' confidence score threshold for a class, which balances the trade-off between localization and recall-precision. In our experiments, we show that, for state-of-the-art object (SOTA) detectors, Optimal LRP provides richer and more discriminative information than AP. We also demonstrate that the best confidence score thresholds vary significantly among classes and detectors. Moreover, we present LRP results of a simple online video object detector which uses a SOTA still image object detector and show that the class-specific optimized thresholds increase the accuracy against the common approach of using a general threshold for all classes. At https://github.com/cancam/LRP we provide the source code that can compute LRP for the PASCAL VOC and MSCOCO datasets. Our source code can easily be adapted to other datasets as well.