Few-Shot Object Detection by Attending to Per-Sample-Prototype

TL;DR

This paper introduces a novel meta-learning approach for few-shot object detection that uses attention mechanisms to treat each support sample as an individual prototype, improving performance especially with diverse support data.

Contribution

The method uniquely employs attention to utilize each support sample as a separate prototype, enhancing few-shot detection performance over traditional averaging methods.

Findings

Effective on PASCAL VOC and COCO benchmarks.

Performance improves with increased diversity among support samples.

Complementary to existing methods, easy to integrate.

Abstract

Few-shot object detection aims to detect instances of specific categories in a query image with only a handful of support samples. Although this takes less effort than obtaining enough annotated images for supervised object detection, it results in a far inferior performance compared to the conventional object detection methods. In this paper, we propose a meta-learning-based approach that considers the unique characteristics of each support sample. Rather than simply averaging the information of the support samples to generate a single prototype per category, our method can better utilize the information of each support sample by treating each support sample as an individual prototype. Specifically, we introduce two types of attention mechanisms for aggregating the query and support feature maps. The first is to refine the information of few-shot samples by extracting shared information between the support samples through attention. Second, each support sample is used as a class code to leverage the information by comparing similarities between each support feature and query features. Our proposed method is complementary to the previous methods, making it easy to plug and play for further improvement. We have evaluated our method on PASCAL VOC and COCO benchmarks, and the results verify the effectiveness of our method. In particular, the advantages of our method are maximized when there is more diversity among support data.