GLE-Net: A Global and Local Ensemble Network for Aerial Object Detection

Jiajia Liao,Yundong Wu,Yujun Liu,Yingchao Piao,Jinhe Su,Guorong Cai

doi:10.1007/s44196-021-00056-3

Abstract

Recent advances in camera-equipped drone applications increased the demand for visual object detection algorithms with deep learning for aerial images. There are several limitations in accuracy for a single deep learning model. Inspired by ensemble learning can significantly improve the generalization ability of the model in the machine learning field, we introduce a novel integration strategy to combine the inference results of two different methods without non-maximum suppression. In this paper, a global and local ensemble network (GLE-Net) was proposed to increase the quality of predictions by considering the global weights for different models and adjusting the local weights for bounding boxes. Specifically, the global module assigns different weights to models. In the local module, we group the bounding boxes that corresponding to the same object as a cluster. Each cluster generates a final predict box and assigns the highest score in the cluster as the score of the final predict box. Experiments on benchmarks VisDrone2019 show promising performance of GLE-Net compared with the baseline network.

Highlights

Object detection in aerial images has become a challenging and active field in computer vision
There are three special challenges for aerial images as follows: (1) aerial datasets mostly high-resolution images; (2) objects typically have small scales relative to the images, and (3) the object distribution of images is not uniform in large scenes. It is difficult for the general-purpose detector to effectively detect objects of the aerial images and the most recent works focus on aerial images (e.g., Context-Aware Detection Network (CAD-Net) [1] R2CNN [2]), which cannot reach the level where the state-of-the-art object detection methods perform on natural images
We propose a global and local ensemble network to enhance multi-model detection results for aerial image object detection, which can serve as an efficient plug-and-play network in existing scene parsing networks

Summary

Introduction

Object detection in aerial images has become a challenging and active field in computer vision. The integrated machine learning model is a common method to improve models’ capability, which has been used in many scenarios since it combines the decision of multiple models to upgrade the overall performance. These approaches have been effectively employed for improving accuracy in some machine learning tasks, and object detection is not an exception. Our proposed method achieves the case of a lower missing rate and obtains higher accuracy performance

Methods

Results

Conclusion