AircraftCarrier_Dataset

This repository is to download the AircraftCarrier dataset

Author: Kai Kang, Gangming Pang, Xun Zhao, Jiabao Wang, Yang Li.

Last Update: 26/01/2020

CITATION:

If you use AircraftCarrier dataset in your research, please cite:

@ARTICLE {AircraftCarrier_Dataset,
author       = "Kai Kang, Gangming Pang, Xun Zhao, Jiabao Wang, Yang Li",
title        = "A New Benchmark for Instance-level Image Classification",
year         = "2020",
}

1.Dataset Download

The AircraftCarrier Dataset are publicly available, if you agree to the terms of its usage, you can obtain it by sending us a scanned Applicant Form to panggangming16@gmail.com. We will send you an e-mail with a download link.

Note that we make no representations or warranties regarding the AircraftCarrier dataset, including but not limited to warranties of non-infringement or fitness for a particular purpose.

2. Dataset Introduction

The dataset contains the following 20 categories: ``Liaoning (CV16)'', "Cavour (CVH550)", "Giuseppe Garibaldi (CVH551)", "USS Nimitz (CVN68)", "USS Dwight David Eisenhower (CVN69)'', "USS Carl Vinson (CVN70)", "USS Theodore Roosevelt (CVN71)", "USS Abraham Lincoln (CVN72)'', "USS George Washington (CVN73)", "USS John C. Stennis (CVN74)", "USS Harry S. Truman (CVN75)", "USS Ronald Reagan (CVN76)'', "USS George H.W. Bush (CVN77)", "USS Gerald R. Ford (CVN78)", "Juan Carlos I (L61)", "HMS Queen Elizabeth (R08)", "INS Viraat (R22)", "Charles de Gaulle (R91)", "HTMS Chakri Naruebet (R911)" and "Admiral Flota Sovetskogo Soyuza Kuznetsov (RN063)''. The following Figure presents examples of the 20 aircraft carriers.

3. Code Usage

Notes: All of the codes we support work well dependent on Python3.6 & Pytorch 1.1.

Before running the code, you should download the dataset and modify the dataset path.

Training

./InstanceCLS_train.sh

Testing

python InstanceCLS_test.py

wenjuntan / AircraftCarrier_Dataset