-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathdataset.html
240 lines (212 loc) · 10 KB
/
dataset.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>ICPR CONTEST</title>
<!-- <link rel="stylesheet" href="http://cdn.static.runoob.com/libs/bootstrap/3.3.7/css/bootstrap.min.css"> -->
<link rel="stylesheet" href="bootstrap-3.3.7-dist/css/bootstrap.min.css">
<link rel="stylesheet" type="text/css" href="css/mystyle.css"
<script src="http://cdn.static.runoob.com/libs/jquery/2.1.1/jquery.min.js"></script>
<!-- <script src="http://cdn.static.runoob.com/libs/bootstrap/3.3.7/js/bootstrap.min.js"></script> -->
<style>
body {
background-color : rgb(250,250,250); /*UCLA bgcolor*/
}
</style>
</head>
<body>
<div class="container">
<div class="jumbotron">
<div class="content">
<h1 style="text-align:center; margin-top:80px; font-weight: bold; color:rgb(100,10,250); font-size: 48px;">
Object Detection in Aerial Images (ODAI)
</h1>
<!-- <p style="text-align:center; color:blue; position:relative; top:7%"> -->
<p style="text-align:center; margin-top:15px; color: rgb(10,10,250); font-size: 32px;">
<strong>
<span style="font-style:italic">A Contest on ICPR'2018 </span>
</strong>
</p>
<!-- <body background="images/largeimage2.jpg"></body> -->
<!-- <img src="images/largeimage2.jpg" class="img-responsive center-block" /> -->
</div>
</div>
<div class="row">
<div class="span12">
<div class="row">
<div class="span6 offset2">
<ul class="nav nav-tabs">
<li> <a href="index.html">Home</a></li>
<li class="active"> <a href="dataset.html">Dataset</a></li>
<li><a href="tasks.html">Tasks</a></li>
<li><a href="evaluation.html">Evaluation</a></li>
<li><a href="results.html">Results</a></li>
<br />
</ul>
</div>
</div>
<h2 style="text-align:left; margin-bottom:10px; margin-top:30px; ">
Image Source and Usage License
</h2>
<p style="text-align:justify;">
The images of in DOTA-v1.0 dataset are manily collected from the Google Earth, some are taken by satellite JL-1,
the others are taken by satellite GF-2 of the China Centre for Resources Satellite Data and Application.
</p>
<p style="text-align:justify;">
Use of the images from Google Earth must respect the corresponding terms of use:
<a href="https://www.google.com/permissions/geoguidelines.html">"Google Earth" terms of use</a>.
</p>
<p style="text-align:justify;">
All images and their associated annotations in DOTA <strong style="color:blue">can be used for academic purposes only, but any commercial use is prohibited</strong>.
</p>
<h2>
Object Category
</h2>
<p>
The object categories in DOTA-v1.0 include: plane, ship, storage tank,
baseball diamond, tennis court, basketball court, ground track field, harbor, bridge, large vehicle,
small vehicle, helicopter, roundabout, soccer ball field and basketball court.
</p>
<!-- <p>
Fifteen categories are chosen and annotated in our dataset
<ul>
<li>plane</li>
<li>ship</li>
<li>storage tank</li>
<li>baseball diamond</li>
<li>tennis court</li>
<li>basketball court</li>
<li>ground track field</li>
<li>harbor</li>
<li>bridge</li>
<li>large vehicle</li>
<li>small vehicle</li>
<li>helicopter</li>
<li>roundabout</li>
<li>soccer ball field</li>
<li>basketball court</li>
</ul>
</p> -->
<h2>
Annotation format
</h2>
<p style="text-align:justify; font-size: 17px">
In the dataset, each instance's location is annotated by a quadrilateral bounding boxes, which can
be denoted as <strong style="color:blue"> "x<sub>1</sub>, y<sub>1</sub>, x<sub>2</sub>, y<sub>2</sub>, x<sub>3</sub>, y<sub>3</sub>, x<sub>4</sub>, y<sub>4</sub>" </strong>
where (x<sub>i</sub>, y<sub>i</sub>) denotes the positions of the oriented bounding boxes' vertices in the image.
The vertices are arranged in a clockwise order.
The following is the Visualization of adopted annotation method. The yellow point represents the <strong style="color:blue">starting point</strong>.
which refers to: (a) top left corner of a plane, (b) top left corner of a large vehicle
diamond, (c) the center of sector-shaped baseball.
</p>
<p style="text-align:justify; font-size: 17px">
Except the annotation
of location, category label is assigned for each instance, which comes from one of the above 15
selected categories, and meanwhile a difficult label is provided which indicates whether the instance
is difficult to be detected(1 for difficult, 0 for not difficult). Annotations for an image are saved in a text file with the same file name.
At the first line, 'imagesource'(from GoogleEarth, GF-2 or JL-1) is given. At the second line, ’gsd’(ground sample distance, the physical size of one image pixel, in meters)
is given. <strong style="color:blue">Note if the 'gsd' is missing, it is annotated to be 'null'.</strong> From third line to last line in annotation text file, annotation for each instance is given. The annotation format is:
</p>
<pre>
<code style="font-size:16px;">
'imagesource':imagesource
'gsd':gsd
x<sub>1</sub>, y<sub>1</sub>, x<sub>2</sub>, y<sub>2</sub>, x<sub>3</sub>, y<sub>3</sub>, x<sub>4</sub>, y<sub>4</sub>, category, difficult
x<sub>1</sub>, y<sub>1</sub>, x<sub>2</sub>, y<sub>2</sub>, x<sub>3</sub>, y<sub>3</sub>, x<sub>4</sub>, y<sub>4</sub>, category, difficult
...
</code>
</pre>
<div class="container">
<!-- <div class="row">
<div class="col-md-2 col-md-offset-5">
<p>
visulizationfsfsdfdsffffff,ffffffffff
</p>
</div>
</div> -->
<div class="row">
<div class="col-xs-4 col-md-4">
<div class="container">
<!-- <h2>Rounded Corners</h2> -->
<!-- <img src="http://captain.whu.edu.cn/DOTAweb/images/plane1.jpg" class="img-rounded" alt="Cinque Terre" width="304" height="236"> -->
<img src="http://captain.whu.edu.cn/DOTAweb/images/plane1.jpg" class="img-rounded" alt="Cinque Terre" width="25%">
<!-- <figcaption style="position:absolute; left:180px; font-size:18px;"><strong>(a)</strong></figcaption> -->
<figcaption style="position:absolute; left:44%; font-size:18px;"><strong>(a)</strong></figcaption>
<!-- <p style="text-align:center;position:relative"> a </p> -->
</div>
</div>
<div class="col-xs-4 col-md-4">
<div class="container">
<!-- <h2>Rounded Corners</h2> -->
<!-- <img src="http://captain.whu.edu.cn/DOTAweb/images/rotaterec.jpg" class="img-rounded" alt="Cinque Terre" width="304" height="236"> -->
<img src="http://captain.whu.edu.cn/DOTAweb/images/rotaterec.jpg" class="img-rounded" alt="Cinque Terre" width="25%">
<!-- <figcaption style="position:absolute; left:180px; font-size:18px"> <strong>(b)</strong></figcaption> -->
<figcaption style="position:absolute; left:45%; font-size:18px"> <strong>(b)</strong></figcaption>
<!-- <p style="text-align:center">b </p> -->
</div>
</div>
<div class="col-xs-4 col-md-4">
<div class="container">
<!-- <h2>Rounded Corners</h2> -->
<!-- <img src="http://captain.whu.edu.cn/DOTAweb/images/baseball2.jpg" class="img-rounded" alt="Cinque Terre" width="304" height="236"> -->
<img src="http://captain.whu.edu.cn/DOTAweb/images/baseball2.jpg" class="img-rounded" alt="Cinque Terre" width="25%">
<!-- <p style="text-align:center">c </p> -->
<!-- <p style="text-align:center">c </p> -->
<figcaption style="position:absolute; left:45%; font-size:18px"> <strong>(c)</strong> </figcaption>
</div>
</div>
</div>
</div>
<!-- <img src="instances-DOTA.jpg" class="img-responsive center-block" /> -->
<br>
<h2>
Dataset Split For DOAI-18
</h2>
<p>
<strong style="color:blue">
The DOAI-18 CONTEST's train, validation sets are the same as the DOTA-v1, test images are from DOTA-v1 test sets and
other extra images.
</strong>
</p>
<h2>
Download
</h2>
<p style="text-align:justify; font-size: 17px">
You can download DOTA-v1.0 from either Baidu Drive or Google Drive, according to your network connections.
</p>
<ul>
<!-- <li>
Dota Training Images <a href="#">[Google Drive]</a> <a href="#">[Baidu Drive]</a>
</li>
<li>
Dota Validation Images <a href="#">[Google Drive]</a> <a href="#">[Baidu Drive]</a>
</li>
<li>
Testing images <a href="#">[Google Drive]</a> <a href="#">[Baidu Drive]</a>
</li>
<!-- <a href="http://captain.whu.edu.cn/datasets/FakeDota_train.tar">DOTA_test</a> -->
<!-- <li>
Dota Training Annotations <a href="#">[Google Drive]</a> <a href="#">[Baidu Drive]</a>
</li>
<li>
Dota Validation Annotations <a href="#">[Google Drive]</a> <a href="#">[Baidu Drive]</a>
</li> -->
<li style="font-size:17px">
DOTA-v1.0 on Baidu Drive: <a href="https://pan.baidu.com/s/1kWyRGaz">Training set</a>, <a href="https://pan.baidu.com/s/1qZCoF72">Validation set</a>, <a href="https://pan.baidu.com/s/1i6ly9Id">Testing images</a>
</li>
<li style="font-size:17px">
DOTA-v1.0 on Google Drive: <a href="https://drive.google.com/drive/folders/1gmeE3D7R62UAtuIFOB9j2M5cUPTwtsxK?usp=sharing">Training set</a>, <a href="https://drive.google.com/drive/folders/1n5w45suVOyaqY84hltJhIZdtVFD9B224?usp=sharing">Validation set</a>, <a href="https://drive.google.com/drive/folders/1mYOf5USMGNcJRPcvRVJVV1uHEalG5RPl?usp=sharing">Testing images</a>
</li>
<br>
<br>
</ul>
</div>
</div>
<br>
<div align="center">
<a href="http://www.amazingcounters.com"><img border="0" src="http://cc.amazingcounters.com/counter.php?i=3220030&c=9660403" alt="AmazingCounters.com"></a>
</div>
<br>
</div>
</body>
</html>