Hausdorff Distance Matching with Adaptive Query Denoising for Rotated Detection Transformer

TL;DR

This paper improves rotated object detection by introducing a Hausdorff distance-based matching cost and an adaptive query denoising method, significantly boosting performance on multiple datasets.

Contribution

It proposes a novel Hausdorff distance-based bipartite matching and an adaptive query denoising strategy to enhance rotated DETR detection accuracy.

Findings

Achieved +4.18 AP50 on DOTA-v2.0

Achieved +4.59 AP50 on DOTA-v1.5

Achieved +4.99 AP50 on DIOR-R

Abstract

Detection Transformers (DETR) have recently set new benchmarks in object detection. However, their performance in detecting rotated objects lags behind established oriented object detectors. Our analysis identifies a key observation: the boundary discontinuity and square-like problem in bipartite matching poses an issue with assigning appropriate ground truths to predictions, leading to duplicate low-confidence predictions. To address this, we introduce a Hausdorff distance-based cost for bipartite matching, which more accurately quantifies the discrepancy between predictions and ground truths. Additionally, we find that a static denoising approach impedes the training of rotated DETR, especially as the quality of the detector's predictions begins to exceed that of the noised ground truths. To overcome this, we propose an adaptive query denoising method that employs bipartite matching to selectively eliminate noised queries that detract from model improvement. When compared to models adopting a ResNet-50 backbone, our proposed model yields remarkable improvements, achieving $\textbf{+4.18}$ AP$_{50}$, $\textbf{+4.59}$ AP$_{50}$, and $\textbf{+4.99}$ AP$_{50}$ on DOTA-v2.0, DOTA-v1.5, and DIOR-R, respectively.