Few-Shot Object Detection via Context-Aware Aggregation for Remote Sensing Images

Few-shot object detection methods have made prodigious progress in recent years. However, these methods are designed for optical images at a single scale, which leads to significantly degraded detection performance due to object scale variation of remote sensing images. In this letter, we propose a few-shot object detection method for the problem of scale variation in remote sensing images. More specifically, our model contains two main components: a context-aware pixel aggregation (CPA) that allows the model to adapt to objects at different scales through different scale convolution and a context-aware feature aggregation (CFA) that enhances context awareness to obtain more semantic information through a graph convolution network (GCN). Experiments on the DIOR dataset demonstrate that our model can achieve a satisfying detection performance on remote sensing images, and our model performs significantly better than the state-of-the-art model.

I. Introduction

With the development of convolution networks, object detection has made great progress in recent years. Object detection has been widely used in many real-world applications, such as hazard detection, environmental monitoring, change detection, and urban planning [2]. The traditional object detection methods (see [9], [12]) suffer from a common issue: a large-scale, diverse dataset is required to train a deep neural network model. Obtaining the labeled data is expensive and time-consuming, which imposes serious limitations on the applicability of object detection in real-world applications. In addition, training the model with a few novel class samples leads to an overfitting problem. Therefore, it becomes important to investigate a special method to learn robust detection from a few novel class samples.