<html><head></head><body><div dir="auto">I did a couple of tests using the python API in which I fed my ex if metadata as part of the prompt which helped, but it's not going to work like, say, plant net for identification of plants.<br><br>Regards,<br>Bill </div><br><br><div class="gmail_quote"><div dir="auto">On March 29, 2025 4:57:25 PM EDT, support@hausoos.com wrote:</div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<pre class="k9mail"><div dir="auto">Did you do recognition down to what level of understanding? For example, for plants / flowers, down to species level (of family, or genus), or for mountains in the background down to name and location?<br><br>Kind Regards<br><br><br>Quoting William Allen <dk@ballen.fastmail.fm>:<br><br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 1ex 0.8ex; border-left: 1px solid #729fcf; padding-left: 1ex;"><div dir="auto">I have used ollama with the granite3.2-vision model to do this locally with pretty good preliminary results. Obviously it’s not integrated with digikam, but I think with some prompt tweaking one could get it to produce some usable and searchable content about images without exposing them to the world.<br><br>Bill<br><br>On 29 Mar 2025, at 14:26, Daniel Bauer wrote:<br><br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 1ex 0.8ex; border-left: 1px solid #ad7fa8; padding-left: 1ex;"><div dir="auto">Hi,<br><br>I read now and then here about face recognition. For me personally this is not interesting, as my archive is ordered by persons.<br><br>But what would be really fantastic would be "general image content analysis" using *local* AI.<br><br>Like, for example, searching for images that contain<br>- a group of persons<br>- mountains in the background<br>- a person with a teddybaer<br>- somebody wearing a red shirt<br>- somebody sleeping/laughing/sitting/running<br>etc.<br><br>Of course, google, fakebook, X etc. can analyze the content of images this way, and very precise, but the (poor) tools available for the public only work with uploaded images, which, besides of all the data protection questions, means "giving them away".<br><br>A local AI (without uploading any data to anywhere) with such capabilities would be simply mind-blowing.<br><br>I have no idea if something like this is even possible, of if it is already planned for the future of digikam, but maybe my post here brings somebody to the idea, and who knows?<br><br>Have a nice weekend.<br><br>Daniel<br><br><br><br><br>--<br>Daniel Bauer photographer Basel Málaga<br>Twitter: @Marsfotografo (often explicit nudes)<br><a href="https://www.patreon.com/danielbauer">https://www.patreon.com/danielbauer</a><br><a href="https://www.daniel-bauer.com">https://www.daniel-bauer.com</a> (nudes)<br></div></blockquote></blockquote><div dir="auto"><br><br><br></div></pre></blockquote></div></body></html>