<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Teaching Google To See Images</title>
	<atom:link href="http://searchengineland.com/teaching-google-to-see-images-10920/feed" rel="self" type="application/rss+xml" />
	<link>http://searchengineland.com/teaching-google-to-see-images-10920</link>
	<description>Search Engine Land: Must Read News About Search Marketing &#38; Search Engines</description>
	<lastBuildDate>Mon, 23 Nov 2009 01:49:26 -0500</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.4</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: gary price</title>
		<link>http://searchengineland.com/teaching-google-to-see-images-10920/comment-page-1#comment-1891</link>
		<dc:creator>gary price</dc:creator>
		<pubDate>Fri, 06 Apr 2007 04:33:25 +0000</pubDate>
		<guid isPermaLink="false">http://searchengineland.com/beta/teaching-google-to-see-images-10920.php#comment-1891</guid>
		<description>One more company that I was just reminded of in this space is piXlogic. They are also doing work in video image recognition.

More in this 2003 article:
CIA funds development of photo-checking software
&lt;a href=&quot;http://www.theage.com.au/articles/2003/06/05/1054700310133.html&quot; rel=&quot;nofollow&quot;&gt;http://www.theage.com.au/articles/2003/06/05/1054700310133.html&lt;/a&gt;

and this 2006 article.
&lt;a href=&quot;http://www.directionsmag.com/article.php?article_id=2331&amp;trv=1&quot; rel=&quot;nofollow&quot;&gt;http://www.directionsmag.com/article.php?article_id=2331&amp;trv=1&lt;/a&gt;

Direct to piXlogic
&lt;a href=&quot;http://www.pixlogic.com/&quot; rel=&quot;nofollow&quot;&gt;http://www.pixlogic.com/&lt;/a&gt;
</description>
		<content:encoded><![CDATA[<p>One more company that I was just reminded of in this space is piXlogic. They are also doing work in video image recognition.</p>
<p>More in this 2003 article:<br />
CIA funds development of photo-checking software<br />
<a href="http://www.theage.com.au/articles/2003/06/05/1054700310133.html" rel="nofollow">http://www.theage.com.au/articles/2003/06/05/1054700310133.html</a></p>
<p>and this 2006 article.<br />
<a href="http://www.directionsmag.com/article.php?article_id=2331&#038;trv=1" rel="nofollow">http://www.directionsmag.com/article.php?article_id=2331&#038;trv=1</a></p>
<p>Direct to piXlogic<br />
<a href="http://www.pixlogic.com/" rel="nofollow">http://www.pixlogic.com/</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nathania Johnson</title>
		<link>http://searchengineland.com/teaching-google-to-see-images-10920/comment-page-1#comment-1890</link>
		<dc:creator>Nathania Johnson</dc:creator>
		<pubDate>Fri, 06 Apr 2007 03:11:40 +0000</pubDate>
		<guid isPermaLink="false">http://searchengineland.com/beta/teaching-google-to-see-images-10920.php#comment-1890</guid>
		<description>I&#039;ve used Like.com (by Riya) and I find it to be pretty decent. I think it has huge potential for e-tailers.
</description>
		<content:encoded><![CDATA[<p>I&#8217;ve used Like.com (by Riya) and I find it to be pretty decent. I think it has huge potential for e-tailers.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: gary price</title>
		<link>http://searchengineland.com/teaching-google-to-see-images-10920/comment-page-1#comment-1889</link>
		<dc:creator>gary price</dc:creator>
		<pubDate>Fri, 06 Apr 2007 01:34:49 +0000</pubDate>
		<guid isPermaLink="false">http://searchengineland.com/beta/teaching-google-to-see-images-10920.php#comment-1889</guid>
		<description>Chris:
Here are a few more projects:

+ LTU Tech
&lt;a href=&quot;http://www.ltutech.com/en/&quot; rel=&quot;nofollow&quot;&gt;http://www.ltutech.com/en/&lt;/a&gt;

+ Image-Seek Database of Corbis Royalty Free Imagery Using Both Metadata and CBIR
&lt;a href=&quot;http://corbis.ltutech.com/&quot; rel=&quot;nofollow&quot;&gt;http://corbis.ltutech.com/&lt;/a&gt;

Also, LTU presentations from the 2003 and 2004 Search Engine Meeting provide excellent intros to CBIR:
+ Organising personal pictures with content analysis technology
From the 2004 meeting. PDF file.
&lt;a href=&quot;http://www.infonortics.com/searchengines/sh04/slides/ltu.pdf&quot; rel=&quot;nofollow&quot;&gt;http://www.infonortics.com/searchengines/sh04/slides/ltu.pdf&lt;/a&gt;
+ Finding the Right Image in the Corbis Collection
From the 2003 meeting. PDF file.
&lt;a href=&quot;http://www.infonortics.com/searchengines/sh03/slides/nastar-2003.pdf&quot; rel=&quot;nofollow&quot;&gt;http://www.infonortics.com/searchengines/sh03/slides/nastar-2003.pdf&lt;/a&gt;

+ xcavator from Cognisign
&lt;a href=&quot;http://www.cognisign.com/&quot; rel=&quot;nofollow&quot;&gt;http://www.cognisign.com/&lt;/a&gt;
&lt;a href=&quot;http://www.xcavator.net/&quot; rel=&quot;nofollow&quot;&gt;http://www.xcavator.net/&lt;/a&gt;

+ Freenet.de (in German, use your fave mechanical translation software)
&lt;a href=&quot;http://www.freenet.de/suche/index.html?search_type=picture&amp;adv=1&quot; rel=&quot;nofollow&quot;&gt;http://www.freenet.de/suche/index.html?search_type=picture&amp;adv=1&lt;/a&gt;
The first options allows you to limit to images with specific words &quot;in the image.&quot; The third, to images with a human face.

+ Photo2Search from Microsoft
Cameraphone searching and content image retrieval
&lt;a href=&quot;http://preview.tinyurl.com/qcqtc&quot; rel=&quot;nofollow&quot;&gt;http://preview.tinyurl.com/qcqtc&lt;/a&gt;

+ Vima Technologies
&lt;a href=&quot;http://www.vimatech.com/&quot; rel=&quot;nofollow&quot;&gt;http://www.vimatech.com/&lt;/a&gt;
Demos here
&lt;a href=&quot;http://www.mediabakery.com/&quot; rel=&quot;nofollow&quot;&gt;http://www.mediabakery.com/&lt;/a&gt;
Look for the &quot;eye&quot; icon below image.
Also, here:
&lt;a href=&quot;http://www.danitadelimont.com/search.asp&quot; rel=&quot;nofollow&quot;&gt;http://www.danitadelimont.com/search.asp&lt;/a&gt;

Cydral (Coming Soon?)
&lt;a href=&quot;http://www.cydral.com/&quot; rel=&quot;nofollow&quot;&gt;http://www.cydral.com/&lt;/a&gt;
</description>
		<content:encoded><![CDATA[<p>Chris:<br />
Here are a few more projects:</p>
<p>+ LTU Tech<br />
<a href="http://www.ltutech.com/en/" rel="nofollow">http://www.ltutech.com/en/</a></p>
<p>+ Image-Seek Database of Corbis Royalty Free Imagery Using Both Metadata and CBIR<br />
<a href="http://corbis.ltutech.com/" rel="nofollow">http://corbis.ltutech.com/</a></p>
<p>Also, LTU presentations from the 2003 and 2004 Search Engine Meeting provide excellent intros to CBIR:<br />
+ Organising personal pictures with content analysis technology<br />
From the 2004 meeting. PDF file.<br />
<a href="http://www.infonortics.com/searchengines/sh04/slides/ltu.pdf" rel="nofollow">http://www.infonortics.com/searchengines/sh04/slides/ltu.pdf</a><br />
+ Finding the Right Image in the Corbis Collection<br />
From the 2003 meeting. PDF file.<br />
<a href="http://www.infonortics.com/searchengines/sh03/slides/nastar-2003.pdf" rel="nofollow">http://www.infonortics.com/searchengines/sh03/slides/nastar-2003.pdf</a></p>
<p>+ xcavator from Cognisign<br />
<a href="http://www.cognisign.com/" rel="nofollow">http://www.cognisign.com/</a><br />
<a href="http://www.xcavator.net/" rel="nofollow">http://www.xcavator.net/</a></p>
<p>+ Freenet.de (in German, use your fave mechanical translation software)<br />
<a href="http://www.freenet.de/suche/index.html?search_type=picture&#038;adv=1" rel="nofollow">http://www.freenet.de/suche/index.html?search_type=picture&#038;adv=1</a><br />
The first options allows you to limit to images with specific words &#8220;in the image.&#8221; The third, to images with a human face.</p>
<p>+ Photo2Search from Microsoft<br />
Cameraphone searching and content image retrieval<br />
<a href="http://preview.tinyurl.com/qcqtc" rel="nofollow">http://preview.tinyurl.com/qcqtc</a></p>
<p>+ Vima Technologies<br />
<a href="http://www.vimatech.com/" rel="nofollow">http://www.vimatech.com/</a><br />
Demos here<br />
<a href="http://www.mediabakery.com/" rel="nofollow">http://www.mediabakery.com/</a><br />
Look for the &#8220;eye&#8221; icon below image.<br />
Also, here:<br />
<a href="http://www.danitadelimont.com/search.asp" rel="nofollow">http://www.danitadelimont.com/search.asp</a></p>
<p>Cydral (Coming Soon?)<br />
<a href="http://www.cydral.com/" rel="nofollow">http://www.cydral.com/</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Silver</title>
		<link>http://searchengineland.com/teaching-google-to-see-images-10920/comment-page-1#comment-1888</link>
		<dc:creator>Silver</dc:creator>
		<pubDate>Thu, 05 Apr 2007 23:39:33 +0000</pubDate>
		<guid isPermaLink="false">http://searchengineland.com/beta/teaching-google-to-see-images-10920.php#comment-1888</guid>
		<description>One aspect worth mentioning would be that Google&#039;s Image Labeler program could easily be used to provide the data necessary to train the system in this Supervised Multiclass Labeling method.

While SML couldn&#039;t really be used for huge numbers of images in real-time for searching, it could be used to process images over time in order to build up the metadata to associate with the images -- providing a rich new source of keyword content.

So, the potential for this to begin functioning for Google&#039;s Image Search is actually quite high.
</description>
		<content:encoded><![CDATA[<p>One aspect worth mentioning would be that Google&#8217;s Image Labeler program could easily be used to provide the data necessary to train the system in this Supervised Multiclass Labeling method.</p>
<p>While SML couldn&#8217;t really be used for huge numbers of images in real-time for searching, it could be used to process images over time in order to build up the metadata to associate with the images &#8212; providing a rich new source of keyword content.</p>
<p>So, the potential for this to begin functioning for Google&#8217;s Image Search is actually quite high.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
