-
Notifications
You must be signed in to change notification settings - Fork 4
/
tasks.html
207 lines (177 loc) · 7.06 KB
/
tasks.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
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>DOTA</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="https://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>-->
<script src="https://cdn.staticfile.org/twitter-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="row">
<div class="span12">
<div id="page-heading" class="jumbotron" style="margin-left: 0px; margin-right: 0px;">
<!-- <div class="content"> -->
<div id="heading-text" class="container">
<!-- <h1 style="text-align:center; margin-top:60px; font-weight: bold;font-size:50">
<strong style="color:red;text-shadow: 0 0 2px black;">DOTA 2</strong>
</h1> -->
<p class="big-text" style="">
<span class="lamem-text" style="margin-right: 2px;font-size: 60px">DOTA</span>
</p>
<p style="font-size: 31px;font-family: 'Arial Narrow'">
<strong>A Large-Scale Benchmark and Challenges for Object Detection in Aerial Images</strong>
</p>
</div>
</div>
<div class="row">
<div class="span4">
</div>
<div class="span4">
<ul class="nav nav-tabs">
<!-- <li class="active"> -->
<li >
<a href="index.html">Home</a>
</li>
<li>
<a href="dataset.html">Dataset</a>
</li>
<li>
<a href="code.html">Code</a>
</li>
<li class="active">
<!-- <li> -->
<a href="tasks.html">Tasks</a>
</li>
<li>
<a href="evaluation.html">Evaluation</a>
</li>
<li class="dropdown">
<a class="dropdown-toggle" data-toggle="dropdown" href="#">
Results <span class="caret"></span>
</a>
<ul class="dropdown-menu">
<li><a href="results.html"> <i class="glyphicon glyphicon-th-list"></i> DOTA-v1.0</a></li>
<li><a href="results_dota15.html"> <i class="glyphicon glyphicon-th-list"></i> DOTA-v1.5</a></li>
<li><a href="results_dota2.html"> <i class="glyphicon glyphicon-th-list"></i> DOTA-v2.0</a></li>
</ul>
</li>
<li>
<a href="people.html">People</a>
</li>
<li>
<a href="external.html">External</a>
</li>
<br />
</ul>
</div>
<div class="span4">
</div>
</div>
<h2>
Overview
</h2>
<p>
We introduce two detection tasks. Task1 uses the initial oriented bounding boxes (OBB) as ground truth.
Task2 uses the generated HBB as ground truth.
Task1 has more practical value and we recommond you to test your algorithms in Task1.
</p>
<h2>Task1 - Detection with oriented bounding boxes <strong>(Recommended)</strong> </h2>
<p>
The aim of this task is to locate the ground object instances with an OBB.
</p>
<!-- <h3>
Evaluation Server
</h3>
<p>
For evaluation, you must registrate and submit on the
<a href="http://www.icdar2017chinese.site:5080/evaluation1/">Evaluation Server</a>
</p> -->
<h3>
Submission Format
</h3>
<p>
You need to submit a zip file containing results for all test images for evaluation. The results are stored
in n (n is the numer of categories) files, <strong style="color:blue">"Task1_plane.txt, Task1_storage-tank.txt, ..."</strong>, each file contains all the detections for a specific category.
Each file is in the following format:
</p>
<pre>
<code style="font-size:16px">
imgname score 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>
imgname score 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>
...
</code>
</pre>
<a href="submissionformat/example_task1.zip", style="font-size: 16px">An example submission of task1 for DOTA-v1.0</a>
<h3>
Evaluation Protocol
</h3>
<p>
The evaluation protocol for oriented bounding box is a little different from the protocol in the
original PASCAL VOC. We use the intersection over the union area of two polygons(ground truth
and prediction) to calculate the IoU. The rest follows the PASCAL VOC.
</p>
<h2>
Task2 - Detection with horizontal bounding boxes
</h2>
<p>
Detecting object with horizontal bounding boxes is usual in many previous contests for object
detection. The aim of this task is to accurately localize the instance in terms of horizontal bounding
box with (xmin, ymin, xmax, ymax) format. In the task, the ground truths for training and testing are generated
by calculating the horizontal bounding boxes over original annotated bounding boxes.
</p>
<p>
</p>
<!-- <h3>
Evaluation Server
</h3>
<p>
For evaluation, you must registrate and submit on the
<a href="http://www.icdar2017chinese.site:5080/evaluation1/">Evaluation Server</a>
</p> -->
<h3>
Submission Format
</h3>
<p>
You need to submit a zip file containing results for all test images for evaluation. The results are stored
in n (n is the numer of categories) files, <strong style="color:blue">"Task2_plane.txt, Task2_storage-tank.txt, ..."</strong>, each file contains all the results for a specific category.
</p>
<pre>
<code style="font-size:16px">
imgname score xmin ymin xmax ymax
imgname score xmin ymin xmax ymax
...
</code>
</pre>
<a href="submissionformat/example_task2.zip", style="font-size: 16px">An example submission of task2 for DOTA-v1.0</a>
<h3>
Evaluation Protocol
</h3>
<p>
The evaluation protocol for horizontal bounding boxes follows the PASCAL VOC benchmark,
which uses mean Average Precision(<strong>mAP</strong>) as metric.
</p>
</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> -->
<!-- <p align="center">-->
<!-- <a href="https://clustrmaps.com/site/1ayvp" title="Visit tracker"><img src="//www.clustrmaps.com/map_v2.png?d=Vs7kEm-F-yc_I1yR89KMrjHwPuM3LWsU-RKxnqEQh3s&cl=ffffff" /></a>-->
<!-- </p>-->
<script type='text/javascript' id='clustrmaps' src='//cdn.clustrmaps.com/map_v2.js?cl=ffffff&w=300&t=tt&d=Vs7kEm-F-yc_I1yR89KMrjHwPuM3LWsU-RKxnqEQh3s'></script>
<br>
</div>
</body>
</html>