<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><font size="-1"><font face="Noto Sans">Hi Serkan,</font></font></p>
<p><font size="-1"><font face="Noto Sans">I am in the process of
extending Mapswipe (<a class="moz-txt-link-freetext" href="http://www.missingmaps.org/">http://www.missingmaps.org/</a>) an app that
is used to create some of the task manager task geometries.
Together with Benni Herfort at Heidelberg university, we're
setting up a workflow to help people validate building
footprints produced by ML systems. Our first target is a
dataset of buildings in Tanzania that contains about 18
million footprints. The aim is to do a bit of preprocessing,
to put them in a decent shape, then have volunteers validate
them against satellite imagery, so that they can then be
imported onto the map. Our hope is that this speeds up mapping
of buildings (and possibly other elements) while also raising
the resulting quality.<br>
</font></font></p>
<p><font size="-1"><font face="Noto Sans">I think this might tie
into what you're doing in 2 ways:</font></font></p>
<p><font size="-1"><font face="Noto Sans">- our output will be a
high quality dataset of validated building footprints that you
could use for training<br>
</font></font></p>
<p><font size="-1"><font face="Noto Sans">- your output could feed
into the app for validation</font></font></p>
<p><font size="-1"><font face="Noto Sans">Happy to tell you more
about what we're doing in a separate discussion, if you think
it's helpful.</font></font></p>
<p><font size="-1"><font face="Noto Sans">Laurent</font></font></p>
<p><font size="-1"><font face="Noto Sans"><br>
</font></font></p>
<div class="moz-cite-prefix">On 20/11/2018 03:44, Serkan Karakulak
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAJ_BQi_cNrhgCmaE7dyj9ScdyTQF6edC7qEoVm0HGxLRMcB7Bg@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">
<div dir="ltr">Hi,
<div>I am a graduate student in a Data Science program, and I
wanted to get in touch with you because me and my two other
friends are interested in working on a machine learning
project to map areas using their satellite images and
produce their labels. If we obtain a high accuracy, we
thought it could be of use to the hotosm and the missingmaps
initiatives.<br>
</div>
<div>
<div><br>
</div>
<div>I have first heard about Missing Maps when I came
across a mapathlon event two years ago in Istanbul. We
were in search for a class project for one of our courses
and I remembered about the mapathlon event and we would be
very happy if we could contribute to the this wonderful
initiative. There are already some previous successful
works on this subject, so we are very hopeful that we
could come up with an algorithm with high accuracy.</div>
<div><br>
</div>
<div>In order to develop the model, we need to extract
bounding boxes of the validated tiles we see at hotosm.
Then we will extract the satellite images and their labels
using the label-maker API which is developed by DevSeed.
Is there a method to extract the coordinates of these
validated tiles? </div>
<div><br>
</div>
<div>Thank you for your time and help.</div>
<div><br>
</div>
<div>Best,</div>
<div><br>
</div>
<div>Serkan</div>
</div>
</div>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
HOT mailing list
<a class="moz-txt-link-abbreviated" href="mailto:HOT@openstreetmap.org">HOT@openstreetmap.org</a>
<a class="moz-txt-link-freetext" href="https://lists.openstreetmap.org/listinfo/hot">https://lists.openstreetmap.org/listinfo/hot</a>
</pre>
</blockquote>
</body>
</html>