<?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/"
	xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: Can Google Solve the Image Search Problem?</title>
	<atom:link href="http://thomashawk.com/2008/04/can-google-solve-image-search-problem.html/feed" rel="self" type="application/rss+xml" />
	<link>http://thomashawk.com/2008/04/can-google-solve-image-search-problem.html</link>
	<description></description>
	<lastBuildDate>Sat, 20 Mar 2010 20:33:33 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.1</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Kurazaybo</title>
		<link>http://thomashawk.com/2008/04/can-google-solve-image-search-problem.html/comment-page-1#comment-2147</link>
		<dc:creator>Kurazaybo</dc:creator>
		<pubDate>Tue, 06 May 2008 15:41:00 +0000</pubDate>
		<guid isPermaLink="false">http://clients.emmense.com/thomashawk/?p=404#comment-2147</guid>
		<description>I disagree, Flickr is not relevant as a general image search engine. Maybe the technology behind is, but I would like to see some justification on why that is the case. I think flickr still relies too much in human generated tags.</description>
		<content:encoded><![CDATA[<p>I disagree, Flickr is not relevant as a general image search engine. Maybe the technology behind is, but I would like to see some justification on why that is the case. I think flickr still relies too much in human generated tags.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Patrick Dodds</title>
		<link>http://thomashawk.com/2008/04/can-google-solve-image-search-problem.html/comment-page-1#comment-2148</link>
		<dc:creator>Patrick Dodds</dc:creator>
		<pubDate>Mon, 28 Apr 2008 13:20:00 +0000</pubDate>
		<guid isPermaLink="false">http://clients.emmense.com/thomashawk/?p=404#comment-2148</guid>
		<description>Thanks for the update on Riya Thomas - I got burned with that too, getting all excited and uploading hundreds of wedding pictures only to find that the software didn&#039;t work at all and that all one was doing was simply labeling the pictures oneself. Sad it didn&#039;t work mind - I had visions of all sorts of possibilities for facial recognition that worked (some of them not so pleasant governmental anti-terrst measures of course, but others quite useful).&lt;br /&gt;BTW, your captcha still isn&#039;t working the first time round (I have the same problem on my blog - v. annoying).</description>
		<content:encoded><![CDATA[<p>Thanks for the update on Riya Thomas &#8211; I got burned with that too, getting all excited and uploading hundreds of wedding pictures only to find that the software didn&#8217;t work at all and that all one was doing was simply labeling the pictures oneself. Sad it didn&#8217;t work mind &#8211; I had visions of all sorts of possibilities for facial recognition that worked (some of them not so pleasant governmental anti-terrst measures of course, but others quite useful).<br />BTW, your captcha still isn&#8217;t working the first time round (I have the same problem on my blog &#8211; v. annoying).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eric</title>
		<link>http://thomashawk.com/2008/04/can-google-solve-image-search-problem.html/comment-page-1#comment-2149</link>
		<dc:creator>Eric</dc:creator>
		<pubDate>Mon, 28 Apr 2008 11:52:00 +0000</pubDate>
		<guid isPermaLink="false">http://clients.emmense.com/thomashawk/?p=404#comment-2149</guid>
		<description>I imagine that the &quot;image labeler&quot; game wasn&#039;t ever really intended to solve the problem - Google probably just wanted some human powered data to compare their algorithms against.&lt;br /&gt;&lt;br /&gt;I agree that for the moment, Flickr&#039;s approach is probably the best, but it has its flaws too - notably if you happen to be looking for a non-photo image.&lt;br /&gt;&lt;br /&gt;(And it kind of boggles my mind that Yahoo never took this same approach and applied it to web search with Delicious)&lt;br /&gt;&lt;br /&gt;But the holy grail in this domain is image recognition; if Google has made a real breakthrough in this area it could be a big deal.</description>
		<content:encoded><![CDATA[<p>I imagine that the &#8220;image labeler&#8221; game wasn&#8217;t ever really intended to solve the problem &#8211; Google probably just wanted some human powered data to compare their algorithms against.</p>
<p>I agree that for the moment, Flickr&#8217;s approach is probably the best, but it has its flaws too &#8211; notably if you happen to be looking for a non-photo image.</p>
<p>(And it kind of boggles my mind that Yahoo never took this same approach and applied it to web search with Delicious)</p>
<p>But the holy grail in this domain is image recognition; if Google has made a real breakthrough in this area it could be a big deal.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
