Fewer is More: Efficient Object Detection in Large Aerial Images

by   Xingxing Xie, et al.

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 consistent accuracy improvements. On extremely large Gaofen-2 images (29200×27620 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 respectively, without sacrificing the accuracy. Code is available at https://github.com/Ranchosky/OAN.


page 3

page 6

page 10

page 11

page 14


DPAttack: Diffused Patch Attacks against Universal Object Detection

Recently, deep neural networks (DNNs) have been widely and successfully ...

S-Extension Patch: A simple and efficient way to extend an object detection model

While building convolutional network-based systems, the toll it takes to...

Kill Two Birds With One Stone: Boosting Both Object Detection Accuracy and Speed With adaptive Patch-of-Interest Composition

Object detection is an important yet challenging task in video understan...

3D Video Object Detection with Learnable Object-Centric Global Optimization

We explore long-term temporal visual correspondence-based optimization f...

Clustered Object Detection in Aerial Images

Detecting objects in aerial images is challenging for at least two reaso...

DATE: Dual Assignment for End-to-End Fully Convolutional Object Detection

Fully convolutional detectors discard the one-to-many assignment and ado...

Scale Match for Tiny Person Detection

Visual object detection has achieved unprecedented ad-vance with the ris...

Please sign up or login with your details

Forgot password? Click here to reset