-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathbadlad.html
411 lines (361 loc) · 18.8 KB
/
badlad.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
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
<!DOCTYPE html>
<html>
<head lang="en">
<meta charset="UTF-8">
<meta http-equiv="x-ua-compatible" content="ie=edge">
<title>BaDLAD: A Large Multi-Domain Bengali
Document Layout Analysis Dataset</title>
<meta name="description" content="">
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta property="og:image" content="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/badlad-icon.png">
<meta property="og:image:type" content="image/png">
<meta property="og:image:width" content="1296">
<meta property="og:image:height" content="840">
<meta property="og:type" content="website" />
<meta property="og:url" content="https://jonbarron.info/zipnerf/"/>
<meta property="og:title" content="BaDLAD: A Large Multi-Domain Bengali
Document Layout Analysis Dataset" />
<meta property="og:description" content="While strides have been made in deep learning based Bengali Optical Character Recognition (OCR) in the past decade, absence
of large Document Layout Analysis (DLA) datasets has hindered the application of OCR in document transcription, e.g., transcribing historical
documents and newspapers. Moreover, rule-based DLA systems that are
currently being employed in practice are not robust to domain variations
and out-of-distribution layouts. To this end, we present the first multidomain large Bengali Document Layout Analysis Dataset: BaDLAD.
This dataset contains 33
, 695 human annotated document samples from
six domains - i) books and magazines ii) public domain govt. documents
iii) liberation war documents iv) new newspapers v) historical newspapers and vi) property deeds; with 710K polygon annotations for four
unit types: text-box, paragraph, image, and table. Through preliminary
experiments benchmarking the performance of existing state-of-the-art
deep learning architectures for English DLA, we demonstrate the efficacy of our dataset in training deep learning based Bengali document
digitization models." />
<meta name="twitter:card" content="summary_large_image" />
<meta name="twitter:title" content="BaDLAD: A Large Multi-Domain Bengali
Document Layout Analysis Dataset" />
<meta name="twitter:description" content="While strides have been made in deep learning based Bengali Optical Character Recognition (OCR) in the past decade, absence
of large Document Layout Analysis (DLA) datasets has hindered the application of OCR in document transcription, e.g., transcribing historical
documents and newspapers. Moreover, rule-based DLA systems that are
currently being employed in practice are not robust to domain variations
and out-of-distribution layouts. To this end, we present the first multidomain large Bengali Document Layout Analysis Dataset: BaDLAD.
This dataset contains 33
, 695 human annotated document samples from
six domains - i) books and magazines ii) public domain govt. documents
iii) liberation war documents iv) new newspapers v) historical newspapers and vi) property deeds; with 710K polygon annotations for four
unit types: text-box, paragraph, image, and table. Through preliminary
experiments benchmarking the performance of existing state-of-the-art
deep learning architectures for English DLA, we demonstrate the efficacy of our dataset in training deep learning based Bengali document
digitization models." />
<meta name="twitter:image" content="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/badlad-icon.png" />
<link rel="icon" href="data:image/svg+xml,<svg xmlns=%22http://www.w3.org/2000/svg%22 viewBox=%220 0 100 100%22><text y=%22.9em%22 font-size=%2290%22>⚡</text></svg>">
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.5/css/bootstrap.min.css">
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.4.0/css/font-awesome.min.css">
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/codemirror/5.8.0/codemirror.min.css">
<link rel="stylesheet" href="css/app.css">
<link rel="stylesheet" href="css/bootstrap.min.css">
<script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.3/jquery.min.js"></script>
<script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.5/js/bootstrap.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/codemirror/5.8.0/codemirror.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/1.5.3/clipboard.min.js"></script>
<script src="js/app.js"></script>
<script src="js/video_comparison.js"></script>
</head>
<body>
<div class="container" id="main">
<div class="row">
<h2 class="col-md-12 text-center">
<b>BaDLAD</b>: A Large Multi-Domain Bengali
Document Layout Analysis Dataset</br>
<small>
ICDAR 2023
</small>
</h2>
</div>
<div class="row">
<div class="col-md-12 text-center">
<ul class="list-inline">
<li>
<a href="https://scholar.google.com/citations?user=JmTaGcYAAAAJ&hl=en">
Shihab Istiaq
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/rakibulranak/">
Rakibul Ranak
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/emon-swe-sust/">
Mahfuzur Emon
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/mobassir-hossen-916a83137/">
Syed Mobassir
</a>
</li>
<li>
<a href="https://mnansary.github.io/">
Nazmuddoha Ansary
</a>
</li>
<li>
<a href="https://scholar.google.com/citations?user=Ks5-xygAAAAJ&hl=en">
Intesur Ahmed
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/fazle-rakib/">
Fazle Rakib
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/shahriardhruvo/?originalSubdomain=bd">
Shahrier Dhrubo
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/souhardya-saha/">
Souhardya Dip
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/akibhasanpavel/?originalSubdomain=bd">
Akib Pavel
</a>
</li>
<li>
<a href="https://www.linkedin.com/in/marsia-haque-meghla-321021207/">
Marsia Meghla
</a>
</li>
<li>
<a href="https://scholar.google.com/citations?user=HaI-oFUAAAAJ&hl=en">
Rezwanul Haque
</a>
</li>
<li>
<a href="https://www.sust.edu/institutes/iict/faculty/10">
Sayma Chowdhury
</a>
</li>
<li>
<a href="https://people.bengali.ai/farig/">
Farig Sadeq
</a>
</li>
<li>
<a href="https://scholar.google.com/citations?user=W_okWy0AAAAJ&hl=en">
Tahsin Reasat
</a>
</li>
<li>
<a href="https://imtiazhumayun.github.io/">
Ahmed Imtiaz Humayun
</a>
</li>
<li>
<a href="https://people.bengali.ai/sushmit/">
Asif Sushmit
</a>
</li>
</br>Bengali.AI
</ul>
</div>
</div>
<div class="row">
<div class="col-md-4 col-md-offset-4 text-center">
<ul class="nav nav-pills nav-justified">
<li>
<a href="https://bengaliai.github.io/">
<image src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/logoBengaliai.jpg" height="60px">
<h4><strong>Bengali.AI Home</strong></h4>
</a>
</li>
<li>
<a href="https://arxiv.org/pdf/2303.05325.pdf">
<image src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/badladpaper.png" height="60px">
<h4><strong>Paper</strong></h4>
</a>
</li>
<li>
<a href="https://github.com/BengaliAI/BADLAD">
<image src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/GitHub-Mark.png" height="60px">
<h4><strong>Github Repo</strong></h4>
</a>
</li>
<li>
<a href="https://www.kaggle.com/datasets/reasat/badlad-train">
<image src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/download.png" height="60px">
<h4><strong>Download Dataset</strong></h4>
</a>
</li>
<li>
<a href="https://www.kaggle.com/datasets/reasat/badlad-train/settings">
<image src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/badladsample.jpg" height="60px">
<h4><strong>4 Million Unannotated Samples</strong></h4>
</a>
</li>
<li>
<a href="https://www.kaggle.com/c/dlsprint2">
<image src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/kaggle.png" height="60px">
<h4><strong>DLSprint2.0 Kaggle Competition</strong></h4>
</a>
</li>
</ul>
</div>
</div>
<!-- <div class="row">
<div class="col-md-8 col-md-offset-2">
<video id="v0" width="100%" autoplay loop muted controls>
<source src="img/teaser.mp4" type="video/mp4" />
</video>
</div>
</div> -->
<div class="row">
<div class="col-md-8 col-md-offset-2">
<div class="image">
<img src="https://raw.githubusercontent.com/BengaliAI/bengaliai.github.io/main/images/badlad.png" alt="Your Image", width = 400, height=320>
</div>
<div class="text">
<p>Predictions of M-RCNN-101 model on BaDLAD Test samples. The con-
tents of the third sample (from the Property deeds domain) has been redacted
for confidentiality. The first 3 samples show only bounding box predictions and
the rest show segmentation boundaries.</p>
</div>
</div>
</div>
<div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Abstract
</h3>
<p class="text-justify">
While strides have been made in deep learning based Bengali Optical Character Recognition (OCR) in the past decade, absence
of large Document Layout Analysis (DLA) datasets has hindered the application of OCR in document transcription, e.g., transcribing historical
documents and newspapers. Moreover, rule-based DLA systems that are
currently being employed in practice are not robust to domain variations
and out-of-distribution layouts. To this end, we present the first multidomain large Bengali Document Layout Analysis Dataset: BaDLAD.
This dataset contains 33
, 695 human annotated document samples from
six domains - i) books and magazines ii) public domain govt. documents
iii) liberation war documents iv) new newspapers v) historical newspapers and vi) property deeds; with 710K polygon annotations for four
unit types: text-box, paragraph, image, and table. Through preliminary
experiments benchmarking the performance of existing state-of-the-art
deep learning architectures for English DLA, we demonstrate the efficacy of our dataset in training deep learning based Bengali document
digitization models.
</p>
</div>
</div>
<!-- <div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Video
</h3>
<div class="text-center">
<div style="position:relative;padding-top:56.25%;">
<iframe src="https://youtube.com/embed/xrrhynRzC8k" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;"></iframe>
</div>
</div>
</div>
</div> -->
<br>
<!-- <div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
360° Video Flythroughs
</h3>
<div class="text-center">
<div style="position:relative;padding-top:56.25%;">
<iframe src="https://www.youtube.com/embed/videoseries?list=PLzPoYEE6Aw7Jzjek1uEIPnpcTDL3u8tb4" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;"></iframe>
<!-- <iframe src="https://youtube.com/embed/jbE2ri8xEZo" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;"></iframe> -->
</div>
</div>
</div> -->
<br>
<!-- <div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
XY aliasing
</h3>
<div class="video-compare-container" id="xyaliasDiv">
<video class="video" id="xyalias" loop playsinline autoPlay muted src="img/xy_alias_swipe_crf27.mp4" onplay="resizeAndPlay(this)"></video>
<canvas height=0 class="videoMerge" id="xyaliasMerge"></canvas>
</div>
<p class="text-justify">
A naive baseline (left) combining mip-NeRF 360 and Instant NGP results in aliasing as the camera moves laterally. Our full method (right) produces prefiltered renderings that do not flicker or shimmer.
</p>
</div>
</div>
-->
<!-- <div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Z aliasing
</h3>
<video id="v0" width="100%" autoplay loop muted controls>
<source src="img/z_alias_pdf_labeled.m4v" type="video/mp4" />
</video>
<p class="text-justify">
The proposal network used for resampling points along rays in mip-NeRF 360 results in an artifact we refer to as <em>z-aliasing</em>, where foreground content alternately appears and disappears as the camera moves toward or away from scene content. Z-aliasing occurs when the initial set of samples from the proposal network is not dense enough and misses thin structures, such as the chair above. Missed content can not be recovered by later rounds of sampling, since no future samples will be placed at that location along the ray. Our improvements to proposal network supervision result in a prefiltered proposal output that preserves the foreground object for all frames in this sequence. The plots above depict samples along a ray for three rounds of resampling (blue, orange, and green lines), with the y axis showing rendering weight (how much each interval contributes to the final rendered color), as a normalized probability density.
</p>
</div>
</div> -->
<div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Citation
</h3>
<div class="form-group col-md-10 col-md-offset-1">
<textarea id="bibtex" class="form-control" readonly>
@article{shihab2023badlad,
title={BaDLAD: A Large Multi-Domain Bengali Document Layout Analysis Dataset},
author={Shihab, Md and Hossain, Istiak and Hasan, Md and Emon, Mahfuzur Rahman and Hossen, Syed Mobassir and Ansary, Md and Ahmed, Intesur and Rakib, Fazle Rabbi and Dhruvo, Shahriar Elahi and Dip, Souhardya Saha and others},
journal={arXiv preprint arXiv:2303.05325},
year={2023}
}</textarea>
</div>
</div>
</div>
<div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Annotators
</h3>
<p class="text-justify">
Fahim Ahmed, Ikram Shams, Asif Ikbal, Md. Junayeth Bhuiyan, Md. Nazibul Islam, Nowshin Noweer Nisa, Dipannita Das Tondra, Umme Humaiara Samia, Emu Akter, Kamrun Naher Pritha, Al Amin Shawon, Samiur Rahman Anadi, Kazi Md. Ashaduj Jaman
<br><br>
.
</p>
</div>
</div>
</div>
<div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Acknowledgements
</h3>
<p class="text-justify">
We are thankful to Center for Bangladesh Genocide Research - CBGR for sharing some invaluable
historical documents for this dataset. We also thank the Department of Software Engineering
in Shahjalal University of Science and Technology,
for their support.
<br><br>
.
</p>
</div>
</div>
</div>
<div class="row">
<div class="col-md-8 col-md-offset-2">
<h3>
Contact
</h3>
<p class="text-justify">
<br><br>
.
</p>
</div>
</div>
</div>
</body>
</html>