<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">Le dim. 30 mars 2025 à 00:46, Kjetil Kjernsmo <<a href="mailto:kjetil@kjernsmo.net">kjetil@kjernsmo.net</a>> a écrit :<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On lørdag 29. mars 2025 19:56:37 CET Michael Miller wrote:<br>
> The auto-tagging models in digiKam can accomplish a small portion of this by<br>
> identifying objects in images like cars, people, animals, etc (YOLOv11<br>
> models), and can also identify some compositional elements like lakeshores,<br>
> mountains, sand dunes, etc (EfficientNetB7 model). These models are the<br>
> best available at the moment that can be run somewhat effectively on a<br>
> person’s computer.<br>
<br>
So, one thing is the kind of laptops people tend to have on their desktop, but <br>
I assume that quite a few Digikam users might tend to be more power users, and <br>
so have stationary computers available that they could use for heavier loads. <br>
I currently have two such, one which has my files and a MySQL database, but it <br>
could do so much more.<br>
<br>
Could this accelerate the progress if there was a framework (in KDE, for <br>
example), to offload processing of such heavy tasks to a different local box?<br>
<br></blockquote><div><br></div><div>We already use the best one : OpenCV</div><div><br></div><div><a href="https://docs.opencv.org/4.x/d2/d58/tutorial_table_of_content_dnn.html">https://docs.opencv.org/4.x/d2/d58/tutorial_table_of_content_dnn.html</a></div><div><br></div><div>It's fully customizable with AI models and can use accelerated hardwares through openCL.</div><div><br></div><div>KDE does not have any stuff like this.</div><div><br></div><div>Best</div><div><br></div><div>Gilles Caulier</div></div></div>