<div dir="ltr"><div dir="ltr"><div>Hi Mike,</div><div><br></div><div class="gmail_quote gmail_quote_container"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
The only other piece of automated metadata is the Image Quality Sorter.
IQS will analyze the image and rate it by applying a flag (red, yellow,
or green).<br></blockquote><div><br></div><div>Thanks, I haven't seen this one yet so I'll check it out.</div><div> <br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
I’ve been toying with the idea of creating a new “plug-and-play” import
pipeline. You can choose to have the images face scanned and
recognized, auto-tagged, and IQS. It sounds like this might be
something you are interested in, too?</blockquote><div><br></div><div>Yeah,
this definitely interests me. I would like as much automated work as
possible, so I like the idea of configuring it once and then having it
run automatically on import. <br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
If you like the automated metadata now, you’ll love what we have coming in 8.6.0.<br>
1. Completely rewritten face engine - It’s more accurate, simpler to
use, and much faster when using “Work on all processor cores”. In some
tests, it’s been over 2 times faster. Face detection and face
recognition now happen in the same scan (there is still an option for
“recognize only”).<br>
2. All new Auto-Tags models. We’re using new versions of YOLO
(YOLOv11), and we replaced ResNet-50 with ImageNet-B7, which is proven
to be over 10% more accurate. We also added controls so you can set
accuracy thresholds. Finally, we added auto-tagging to the Labels view
in the left sidebar so it’s easier to access.<br></blockquote><div><br></div><div>This all sounds great. Do you know approximately when it will be released? <br></div><div><br></div><div>Thanks!</div><div>Paul</div></div><br></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Tue, Mar 4, 2025 at 11:19 AM Michael Miller <<a href="mailto:michael_miller@msn.com">michael_miller@msn.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi Paul,<br>
Thank you for being a digiKam user! We love to hear when people are using digiKam. It makes the hours spend developing it worthwhile.<br>
<br>
> 1. Is there anything else I'm missing? Are there other ways to auto tag or classify images?<br>
<br>
The only other piece of automated metadata is the Image Quality Sorter. IQS will analyze the image and rate it by applying a flag (red, yellow, or green).<br>
<br>
> 2. Is there a way to run all of this automatically when importing new images? I have digikam set to watch my directory, and I copy files in from other sources. I would love for it to auto scan and tag everything in the background as it auto imports, instead of having to run 3 manual scans (that each take quite a long time).<br>
I’ve been toying with the idea of creating a new “plug-and-play” import pipeline. You can choose to have the images face scanned and recognized, auto-tagged, and IQS. It sounds like this might be something you are interested in, too?<br>
<br>
<br>
If you like the automated metadata now, you’ll love what we have coming in 8.6.0.<br>
1. Completely rewritten face engine - It’s more accurate, simpler to use, and much faster when using “Work on all processor cores”. In some tests, it’s been over 2 times faster. Face detection and face recognition now happen in the same scan (there is still an option for “recognize only”).<br>
2. All new Auto-Tags models. We’re using new versions of YOLO (YOLOv11), and we replaced ResNet-50 with ImageNet-B7, which is proven to be over 10% more accurate. We also added controls so you can set accuracy thresholds. Finally, we added auto-tagging to the Labels view in the left sidebar so it’s easier to access.<br>
<br>
Cheers,<br>
Mike<br>
<br>
> On Mar 4, 2025, at 12:26 PM, Paul Gross <<a href="mailto:pgross@gmail.com" target="_blank">pgross@gmail.com</a>> wrote:<br>
> <br>
> Hi Digikam folks,<br>
> <br>
> I'm a long time user of Digikam, so thank you for this great software! Recently, I've been getting into more tagging and wondering if there is a better way to automate some of this.<br>
> <br>
> Currently, after I import images to Digikam, I do the following:<br>
> - Go to people tab and run a scan to “Detect faces”<br>
> - Then run a scan to “Recognize faces”<br>
> - Then to go Tools -> Maintenance and run “Auto-tags Assignment”<br>
> <br>
> Questions:<br>
> <br>
> 1. Is there anything else I'm missing? Are there other ways to auto tag or classify images?<br>
> <br>
> 2. Is there a way to run all of this automatically when importing new images? I have digikam set to watch my directory, and I copy files in from other sources. I would love for it to auto scan and tag everything in the background as it auto imports, instead of having to run 3 manual scans (that each take quite a long time).<br>
> <br>
> Thanks!<br>
> Paul<br>
<br>
</blockquote></div></div>