<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
Is it beneficial for me to continue getting examples? I can focus on only getting Russian written forms for now and we can do the same thing for Ukrainian if it ends up working out.
<div class=""><br class="">
</div>
<div class="">
<div class="">
<div class="">
<div class="">
<div><font face="OpenSans-Regular" size="2" class="">Luke Sciulli</font><br class="">
<font face="OpenSans-Regular" size="2" class="">Senior Research Analyst | Manager</font><br class="">
<font face="OpenSans-Bold" size="2" color="#c41230" class=""><span style="font-style: normal;" class=""><b class="">Carnegie Mellon University</b></span></font><br class="">
<font face="OpenSans-Regular" size="2" class="">C: 412.216.1864</font></div>
</div>
<div><br class="">
<blockquote type="cite" class="">
<div class="">On Jul 13, 2022, at 10:15 PM, Barnabas Poczos <<a href="mailto:bapoczos@cs.cmu.edu" class="">bapoczos@cs.cmu.edu</a>> wrote:</div>
<br class="Apple-interchange-newline">
<div class="">
<div class="">Hi Artur,<br class="">
<br class="">
My student, Hai Pham (cc-ed), created an OCR system that worked pretty well.<br class="">
We mostly used it for handwritten text, but it can also be used for typed text.<br class="">
<br class="">
It needs to be trained though on a dataset. We used a few CMU<br class="">
undergrads who gladly created training data for us.<br class="">
Please find the paper about this system attached.<br class="">
<br class="">
Best,<br class="">
Barnabas<br class="">
<br class="">
<br class="">
======================<br class="">
Barnabas Poczos, PhD<br class="">
Associate Professor<br class="">
Machine Learning Department<br class="">
Carnegie Mellon University<br class="">
<br class="">
On Wed, Jul 13, 2022 at 7:49 AM Artur Dubrawski <<a href="mailto:awd@cs.cmu.edu" class="">awd@cs.cmu.edu</a>> wrote:<br class="">
<blockquote type="cite" class=""><br class="">
Team,<br class="">
<br class="">
I need guidance regarding OCR capabilities, particularly for the Cirrylic alphabet lettering. If we can read forms filled with handwritten text from images or scans, we could create a societally important research opportunity, expanding our impact in healthcare.<br class="">
<br class="">
Let me know your thoughts please.<br class="">
<br class="">
Cheers<br class="">
Artur<br class="">
<br class="">
</blockquote>
<span id="cid:f_l5kedi030"><2008.08148.pdf></span></div>
</div>
</blockquote>
</div>
<br class="">
</div>
</div>
</div>
</body>
</html>