BandRe: Rethinking Band-Pass Filters for Scale-Wise Object Detection Evaluation

TL;DR

This paper introduces new scale-wise evaluation metrics for object detectors that balance detail and reliability, using a filter bank approach, and demonstrates their effectiveness across methods and datasets.

Contribution

It proposes novel scale-wise metrics employing a filter bank of triangular and trapezoidal filters, improving evaluation reliability and granularity.

Findings

Metrics effectively distinguish method differences

Metrics reveal dataset-specific performance variations

Proposed approach enhances evaluation insights

Abstract

Scale-wise evaluation of object detectors is important for real-world applications. However, existing metrics are either coarse or not sufficiently reliable. In this paper, we propose novel scale-wise metrics that strike a balance between fineness and reliability, using a filter bank consisting of triangular and trapezoidal band-pass filters. We conduct experiments with two methods on two datasets and show that the proposed metrics can highlight the differences between the methods and between the datasets. Code is available at https://github.com/shinya7y/UniverseNet .