Poly Kernel Inception Network for Remote Sensing Detection

TL;DR

This paper introduces PKINet, a novel network for remote sensing object detection that uses multi-scale convolution kernels and a context attention module to improve detection across varied object scales and contexts.

Contribution

The paper proposes PKINet with multi-scale kernels and a context anchor attention module, advancing remote sensing detection performance without dilation-based convolutions.

Findings

Improved detection accuracy on DOTA-v1.0, DOTA-v1.5, HRSC2016, and DIOR-R datasets.

Effective handling of scale variation and contextual diversity in remote sensing images.

Outperforms existing methods in benchmark tests.

Abstract

Object detection in remote sensing images (RSIs) often suffers from several increasing challenges, including the large variation in object scales and the diverse-ranging context. Prior methods tried to address these challenges by expanding the spatial receptive field of the backbone, either through large-kernel convolution or dilated convolution. However, the former typically introduces considerable background noise, while the latter risks generating overly sparse feature representations. In this paper, we introduce the Poly Kernel Inception Network (PKINet) to handle the above challenges. PKINet employs multi-scale convolution kernels without dilation to extract object features of varying scales and capture local context. In addition, a Context Anchor Attention (CAA) module is introduced in parallel to capture long-range contextual information. These two components work jointly to advance the performance of PKINet on four challenging remote sensing detection benchmarks, namely DOTA-v1.0, DOTA-v1.5, HRSC2016, and DIOR-R.