HS-FPN: High Frequency and Spatial Perception FPN for Tiny Object Detection

TL;DR

This paper introduces HS-FPN, a novel network that enhances tiny object detection by emphasizing high frequency features and spatial dependencies, addressing limitations of traditional FPN models.

Contribution

The paper proposes two modules, HFP and SDP, to enrich tiny object features and capture spatial dependencies, improving detection performance.

Findings

HS-FPN outperforms state-of-the-art models on AI-TOD dataset.

High frequency responses effectively highlight tiny object features.

Spatial dependency perception improves detection accuracy.

Abstract

The introduction of Feature Pyramid Network (FPN) has significantly improved object detection performance. However, substantial challenges remain in detecting tiny objects, as their features occupy only a very small proportion of the feature maps. Although FPN integrates multi-scale features, it does not directly enhance or enrich the features of tiny objects. Furthermore, FPN lacks spatial perception ability. To address these issues, we propose a novel High Frequency and Spatial Perception Feature Pyramid Network (HS-FPN) with two innovative modules. First, we designed a high frequency perception module (HFP) that generates high frequency responses through high pass filters. These high frequency responses are used as mask weights from both spatial and channel perspectives to enrich and highlight the features of tiny objects in the original feature maps. Second, we developed a spatial dependency perception module (SDP) to capture the spatial dependencies that FPN lacks. Our experiments demonstrate that detectors based on HS-FPN exhibit competitive advantages over state-of-the-art models on the AI-TOD dataset for tiny object detection.