Gaussian Combined Distance: A Generic Metric for Object Detection

TL;DR

This paper introduces the Gaussian Combined Distance (GCD), a new scale-invariant similarity metric for object detection that improves localization accuracy and generalizes well across datasets, outperforming existing metrics like IoU and Wasserstein Distance.

Contribution

The paper proposes GCD, a novel scale-invariant metric that enables joint optimization of bounding box parameters, leading to enhanced detection performance and faster convergence.

Findings

GCD achieves state-of-the-art results on AI-TOD-v2 for tiny objects.

GCD outperforms Wasserstein Distance on MS-COCO-2017 and Visdrone-2019 datasets.

GCD improves model convergence and localization accuracy.

Abstract

In object detection, a well-defined similarity metric can significantly enhance model performance. Currently, the IoU-based similarity metric is the most commonly preferred choice for detectors. However, detectors using IoU as a similarity metric often perform poorly when detecting small objects because of their sensitivity to minor positional deviations. To address this issue, recent studies have proposed the Wasserstein Distance as an alternative to IoU for measuring the similarity of Gaussian-distributed bounding boxes. However, we have observed that the Wasserstein Distance lacks scale invariance, which negatively impacts the model's generalization capability. Additionally, when used as a loss function, its independent optimization of the center attributes leads to slow model convergence and unsatisfactory detection precision. To address these challenges, we introduce the Gaussian Combined Distance (GCD). Through analytical examination of GCD and its gradient, we demonstrate that GCD not only possesses scale invariance but also facilitates joint optimization, which enhances model localization performance. Extensive experiments on the AI-TOD-v2 dataset for tiny object detection show that GCD, as a bounding box regression loss function and label assignment metric, achieves state-of-the-art performance across various detectors. We further validated the generalizability of GCD on the MS-COCO-2017 and Visdrone-2019 datasets, where it outperforms the Wasserstein Distance across diverse scales of datasets. Code is available at https://github.com/MArKkwanGuan/mmdet-GCD.