<div dir="ltr">Ah thanks for the feedback...<div><br></div><div>Gilles Caulier</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Le ven. 3 janv. 2020 à 09:00, Tóth Csaba <<a href="mailto:ignis@domen.hu">ignis@domen.hu</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">
<div>
<div>Hello!</div>
<div>Its working with this file. I deleted
an already detected faces to look what happen, and its found it.</div>
<div>It was my mistake to try it in a
collection with few usable faces</div>
<div><br>
</div>
<div>Thanx all help</div>
<div>Csaba<br>
</div>
<div><br>
</div>
<div>On 2020-01-03 11:13, Gilles Caulier
wrote:<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<div>Hi Tóth and happy new year.</div>
<div><br>
</div>
<div>See this report <a href="https://bugs.kde.org/show_bug.cgi?id=415561" target="_blank">https://bugs.kde.org/show_bug.cgi?id=415561</a></div>
<div><br>
</div>
<div>Please comment on this file directly and follow the
discussions.</div>
<div><br>
</div>
<div>Best</div>
<div>Gilles Caulier</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">Le jeu. 2 janv. 2020
à 17:54, Tóth Csaba <<a href="mailto:ignis@domen.hu" target="_blank">ignis@domen.hu</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">Hello!<br>
Still trying..<br>
digiKam-7.0.0-beta1-20200102T060110-Win64.exe<br>
I try to set up the<br>
detect faces: "clear unconfirmed results and rescan" with
"detect faces" <br>
and the accuracy 60% with and without "clear and revuld all
training data"<br>
no new result (but there is tons of not detected faces)<br>
and the same settings the "Recognize faces", still no new
result.<br>
<br>
If I run "detect faces" on a picture, what settings will be
used? (not <br>
found faces too)<br>
<br>
Thanx<br>
Csaba<br>
<br>
On 2020-01-01 11:34, Maik Qualmann wrote:<br>
> Perform a new build of the training data by activating
the checkbox and<br>
> starting a run. Then try to recognize faces again, set
the accuracy to a value<br>
> between 50-80.<br>
><br>
> Maik<br>
><br>
> Am Mittwoch, 1. Januar 2020, 11:07:19 CET schrieb Tóth
Csaba:<br>
>> Hello!<br>
>> Today I upgraded to
digiKam-7.0.0-beta1-20191231T172902-Win64 to run the<br>
>> face recognize.<br>
>> The detection look like working well, but no new
Unconfirmed faces<br>
>><br>
>> Thanx<br>
>> Csaba<br>
>><br>
>> On 2019-12-29 16:56, Tóth Csaba wrote:<br>
>>> Hello!<br>
>>> Now the digikam on windows10 at "clear and
rebuild all training data"<br>
>>> simple thinking, and quit.<br>
>>> (digiKam-7.0.0-beta1-20191227T122444-Win32.exe)<br>
>>><br>
>>> Thanx<br>
>>> Csaba<br>
>>><br>
>>> On 2019-12-29 13:42, Gilles Caulier wrote:<br>
>>>> I think this problem was recently found and
fixed by Maik.<br>
>>>><br>
>>>> Please try the current daily bundles
available here :<br>
>>>><br>
>>>> <a href="https://files.kde.org/digikam/" rel="noreferrer" target="_blank">https://files.kde.org/digikam/</a><br>
>>>><br>
>>>> At least, Windows and MacOS version are up
to date. Linux bundle is a<br>
>>>> little bit outdated, i'm not sure if it
include this fix.<br>
>>>><br>
>>>> Gilles Caulier<br>
>>>><br>
>>>> Le dim. 29 déc. 2019 à 13:21, Tóth Csaba
<<a href="mailto:ignis@domen.hu" target="_blank">ignis@domen.hu</a><br>
>>>><br>
>>>> <mailto:<a href="mailto:ignis@domen.hu" target="_blank">ignis@domen.hu</a>>>
a écrit :<br>
>>>> Hello!<br>
>>>> <br>
>>>> I'd like to try out the new face
detection function.<br>
>>>> <br>
>>>> I rescan all faces (many time) form
the people tab ("scan<br>
>>>> collection for<br>
>>>> faces"), and run the recognize faces
too (many times with different<br>
>>>> settings, even with reteach too), but
no new image in the<br>
>>>> "unconfirmed"<br>
>>>> folder.<br>
>>>> <br>
>>>> <br>
>>>> How can I run the new (deep learning)
recognize on the unknown faces?<br>
>>>> <br>
>>>> Thanx<br>
>>>> <br>
>>>> Csaba<br>
><br>
><br>
><br>
<br>
</blockquote>
</div>
</div>
</blockquote>
<p><br>
</p>
</div>
</blockquote></div>