A Comparative Attention Framework for Better Few-Shot Object Detection on Aerial Images

TL;DR

This paper introduces a benchmarking framework for attention-based Few-Shot Object Detection (FSOD) methods on aerial images, revealing performance gaps and proposing a novel multiscale alignment technique to enhance small object detection.

Contribution

It presents a flexible benchmarking framework for attention-based FSOD, compares existing methods, and introduces XQSA, a new multiscale alignment approach that improves small object detection in aerial images.

Findings

FSOD performs worse on aerial images than natural images.

Small objects are a key challenge in few-shot detection.

XQSA significantly outperforms existing methods on DOTA and DIOR datasets.

Abstract

Few-Shot Object Detection (FSOD) methods are mainly designed and evaluated on natural image datasets such as Pascal VOC and MS COCO. However, it is not clear whether the best methods for natural images are also the best for aerial images. Furthermore, direct comparison of performance between FSOD methods is difficult due to the wide variety of detection frameworks and training strategies. Therefore, we propose a benchmarking framework that provides a flexible environment to implement and compare attention-based FSOD methods. The proposed framework focuses on attention mechanisms and is divided into three modules: spatial alignment, global attention, and fusion layer. To remain competitive with existing methods, which often leverage complex training, we propose new augmentation techniques designed for object detection. Using this framework, several FSOD methods are reimplemented and compared. This comparison highlights two distinct performance regimes on aerial and natural images: FSOD performs worse on aerial images. Our experiments suggest that small objects, which are harder to detect in the few-shot setting, account for the poor performance. Finally, we develop a novel multiscale alignment method, Cross-Scales Query-Support Alignment (XQSA) for FSOD, to improve the detection of small objects. XQSA outperforms the state-of-the-art significantly on DOTA and DIOR.