首页|Fewer is more:efficient object detection in large aerial images

Fewer is more:efficient object detection in large aerial images

扫码查看
原文链接
  • 万方数据
  • 维普
Current mainstream object detection methods for large aerial images usually divide large images into patches and then exhaustively detect the objects of interest on all patches,no matter whether there exist objects or not.This paradigm,although effective,is inefficient because the detectors have to go through all patches,severely hindering the inference speed.This paper presents an objectness activation network(OAN)to help detectors focus on fewer patches but achieve more efficient inference and more accurate results,enabling a simple and effective solution to object detection in large images.In brief,OAN is a light fully-convolutional network for judging whether each patch contains objects or not,which can be easily integrated into many object detectors and jointly trained with them end-to-end.We extensively evaluate our OAN with five advanced detectors.Using OAN,all five detectors acquire more than 30.0％speed-up on three large-scale aerial image datasets,meanwhile with consistent accuracy improvements.On extremely large Gaofen-2 images(29200 x 27620 pixels),our OAN improves the detection speed by 70.5％.Moreover,we extend our OAN to driving-scene object detection and 4K video object detection,boosting the detection speed by 112.1％and 75.0％,respectively,without sacrificing the accuracy.

efficient object detectionlarge aerial imagesobjectness activation network

Xingxing XIE、Gong CHENG、Qingyang LI、Shicheng MIAO、Ke LI、Junwei HAN

展开 >

School of Automation,Northwestern Polytechnical University,Xi'an 710072,China

Zhengzhou Institute of Surveying and Mapping,Zhengzhou 450052,China

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaNatural Science Basic Research Program of ShaanxiNatural Science Basic Research Program of ShaanxiFundamental Research Funds for the Central UniversitiesDoctorate Foundation of Northwestern Polytechnical University

62136007623762232021JC-162023-JC-ZD-36CX2021082

2024

10.1007/s11432-022-3718-5

中国科学:信息科学(英文版)
中国科学院

中国科学:信息科学(英文版)

CSTPCDEI
影响因子：0.715
ISSN：1674-733X
年,卷(期)：2024.67(1)
  • 71