<html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /></head><body style='font-size: 10pt; font-family: Verdana,Geneva,sans-serif'>
<p>Nextcloud has some good examples of face matching <a href="https://apps.nextcloud.com/apps/facerecognition">https://apps.nextcloud.com/apps/facerecognition</a> so you can filter and copy out, though you would need to setup the source as external storage moutned on your server. free as in opensource, but you can run it in a container. requires 4GB RAM for the pdlib face recognition elements though, but at least you dont have to upload everything to google.</p>
<p>ACDsee also have a face recognition element you might be interested in, that is an image cataloguing app so you can more easily use it, though it does require an indexing stage and then face recognition stage takes a while. the photo studio is a few hundred bucks though, but the face recogntion works well and can be fine tuned easily.</p>
<p>If you are looking for a simple press 1 to copy directory 1 while you cycle through, check out <a href="http://www.irfanview.com">www.irfanview.com</a> - it even has a thumbnail view . it's free it's fast and it's quick</p>
<p>Would also recommend the disk image approach for later review - especially if there are other friends or family that would like to see/share those same images you choose to discard</p>
<p><br /></p>
<p id="reply-intro">On 2021-06-11 06:35, Dean Bergin wrote:</p>
<blockquote type="cite" style="padding: 0 0.4em; border-left: #1010ff 2px solid; margin: 0">
<div id="replybody1">
<div dir="auto">I wonder if OpenCV template matching could help in this situation?
<div dir="auto"> </div>
<div dir="auto"><a href="https://docs.opencv.org/master/d4/dc6/tutorial_py_template_matching.html" target="_blank" rel="noopener noreferrer">https://docs.opencv.org/master/d4/dc6/tutorial_py_template_matching.html</a></div>
<div dir="auto"> </div>
<div dir="auto">Probably something I will play around with in case I'm faced with a similar problem, but as someone pointed out Google Photos (GP) is pretty good at this so the OpenCV approach seems like reinventing the wheel.</div>
<div dir="auto"> </div>
<div dir="auto">In any case, gaining access to someone's device makes the matter less trivial however, you might need to suggest that person synchronise their camera roll to a service like GP so your job is less of a chore as they can do all the analytics for you.</div>
<div dir="auto"> </div>
<div dir="auto">There's probably APIs for GP to leverage the image processing and suck down those that match a given complex criteria...</div>
</div>
<br />
<div class="v1gmail_quote">
<div class="v1gmail_attr" dir="ltr">On Thu, 10 Jun 2021, 22:29 Brad Campbell, <<a href="mailto:brad@fnarfbargle.com" rel="noreferrer">brad@fnarfbargle.com</a>> wrote:</div>
<blockquote class="v1gmail_quote" style="margin: 0 0 0 .8ex; border-left: 1px #ccc solid; padding-left: 1ex;">On 10/6/21 10:25 pm, Benjamin wrote:<br />> The reason I thought to stress it is it gives you the chance to go back and pick out things you missed or you didn't know were important vs doing it directly. <br />> <br />> You still have the job you're actually trying to do to work on and I don't have easy answers, but treat it the same way you'd treat recovering a database when the recovery tools might not work perfect and in fact you might need to try multiple times.<br />> <br />> It's easy to throw away a drive thinking you pulled everything you cared about then have the "oh f*** I forgot this" moment, especially when you are dealing with so many memories and unique things, and especially when some of the significant things might not be obvious yet to friends and family of the deceased. Little things that can't be recreated once disposed of.<br />> <br />> Some day it gets to the point of diminishing returns and you can't ruminate on it forever, but at least a few runs will be needed regardless of how great a "pick the photos that are important" solution ends up being to use.<br />> <br />> Again, if you're painfully aware of it - sorry.<br />> <br /><br />Nah, all good. My brain is mush so it's an important reminder. Thanks.<br /><br />_______________________________________________<br />PLUG discussion list: <a href="mailto:plug@plug.org.au" rel="noreferrer">plug@plug.org.au</a><br /><a href="http://lists.plug.org.au/mailman/listinfo/plug" target="_blank" rel="noopener noreferrer">http://lists.plug.org.au/mailman/listinfo/plug</a><br />Committee e-mail: <a href="mailto:committee@plug.org.au" rel="noreferrer">committee@plug.org.au</a><br />PLUG Membership: <a href="http://www.plug.org.au/membership" target="_blank" rel="noopener noreferrer">http://www.plug.org.au/membership</a></blockquote>
</div>
</div>
<br />
<div class="pre" style="margin: 0; padding: 0; font-family: monospace">_______________________________________________<br />PLUG discussion list: <a href="mailto:plug@plug.org.au">plug@plug.org.au</a><br /><a href="http://lists.plug.org.au/mailman/listinfo/plug" target="_blank" rel="noopener noreferrer">http://lists.plug.org.au/mailman/listinfo/plug</a><br />Committee e-mail: <a href="mailto:committee@plug.org.au">committee@plug.org.au</a><br />PLUG Membership: <a href="http://www.plug.org.au/membership" target="_blank" rel="noopener noreferrer">http://www.plug.org.au/membership</a></div>
</blockquote>
<p><br /></p>
</body></html>