Boundary and Position Information Mining for Aerial Small Object Detection

TL;DR

This paper introduces BPIM, a novel framework that enhances small object detection in aerial images by integrating boundary, position, and scale information through attention mechanisms and multi-scale feature fusion.

Contribution

The paper proposes a comprehensive boundary and position information mining framework that improves small object detection accuracy in UAV imagery, outperforming baseline models and achieving state-of-the-art results.

Findings

BPIM outperforms baseline Yolov5-P2 on multiple datasets.

BPIM achieves state-of-the-art detection performance with comparable computation.

Boundary and position information integration enhances small object perception.

Abstract

Unmanned Aerial Vehicle (UAV) applications have become increasingly prevalent in aerial photography and object recognition. However, there are major challenges to accurately capturing small targets in object detection due to the imbalanced scale and the blurred edges. To address these issues, boundary and position information mining (BPIM) framework is proposed for capturing object edge and location cues. The proposed BPIM includes position information guidance (PIG) module for obtaining location information, boundary information guidance (BIG) module for extracting object edge, cross scale fusion (CSF) module for gradually assembling the shallow layer image feature, three feature fusion (TFF) module for progressively combining position and boundary information, and adaptive weight fusion (AWF) module for flexibly merging the deep layer semantic feature. Therefore, BPIM can integrate boundary, position, and scale information in image for small object detection using attention mechanisms and cross-scale feature fusion strategies. Furthermore, BPIM not only improves the discrimination of the contextual feature by adaptive weight fusion with boundary, but also enhances small object perceptions by cross-scale position fusion. On the VisDrone2021, DOTA1.0, and WiderPerson datasets, experimental results show the better performances of BPIM compared to the baseline Yolov5-P2, and obtains the promising performance in the state-of-the-art methods with comparable computation load.