<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en-gb">
	<link rel="self" type="application/atom+xml" href="https://blueirissoftware.com/forum/app.php/feed/topic/4259" />

	<title>Blue Iris</title>
	<subtitle>Blue Iris User Group</subtitle>
	<link href="https://blueirissoftware.com/forum/index.php" />
	<updated>2024-07-25T04:54:22+00:00</updated>

	<author><name><![CDATA[Blue Iris]]></name></author>
	<id>https://blueirissoftware.com/forum/app.php/feed/topic/4259</id>

		<entry>
		<author><name><![CDATA[billiesavage]]></name></author>
		<updated>2024-07-25T04:54:22+00:00</updated>

		<published>2024-07-25T04:54:22+00:00</published>
		<id>https://blueirissoftware.com/forum/viewtopic.php?p=18702#p18702</id>
		<link href="https://blueirissoftware.com/forum/viewtopic.php?p=18702#p18702"/>
		<title type="html"><![CDATA[Train Yolo 5.6.2 with Blue Iris alerts]]></title>

		
		<content type="html" xml:base="https://blueirissoftware.com/forum/viewtopic.php?p=18702#p18702"><![CDATA[
I have been running Blue Iris alongside CodeProjct AI (2.3.4) and have had good luck training some custom models (via the CodeProject built in module). Over the last several months I have hundreds alerts/positive AI recognitions that generate .DAT files in Blue Iris along with the images with the AI tags. I am wondering if I can somehow harvest the .DAT files or blue iris alert clips/images in an easy way to further train a custom model within the CodeProject training module. I am hoping there is a way to use the existing alerts/data to avoid having to manually tag all training images with their respective boxes/labels and such. Has anyone done anything similar?<p>Statistics: Posted by <a href="https://blueirissoftware.com/forum/memberlist.php?mode=viewprofile&amp;u=8165">billiesavage</a> — Thu Jul 25, 2024 4:54 am</p><hr />
]]></content>
	</entry>
	</feed>
