<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<?xml-stylesheet href="http://www.blogger.com/styles/atom.css" type="text/css"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <id>http://groups.google.com/group/viper-toolkit</id>
  <title type="text">ViPER Mailing List Google Group</title>
  <subtitle type="text">
  Mailing list for discussion about the University of Maryland LAMP lab&#39;s ViPER toolkit for editing video metadata and evaluating video processing algorithms.
  </subtitle>
  <link href="/group/viper-toolkit/feed/atom_v1_0_msgs.xml" rel="self" title="ViPER Mailing List feed"/>
  <updated>2008-07-16T19:18:21Z</updated>
  <generator uri="http://groups.google.com" version="1.99">Google Groups</generator>
  <entry>
  <author>
  <name>annie</name>
  <email>mittal.an...@gmail.com</email>
  </author>
  <updated>2008-07-16T19:18:21Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/17f242b87d2e0858/84ee3f745fd028f4?show_docid=84ee3f745fd028f4</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/17f242b87d2e0858/84ee3f745fd028f4?show_docid=84ee3f745fd028f4"/>
  <title type="text">Making the text file for attribute of descriptor from XML output file of VIPER</title>
  <summary type="html" xml:space="preserve">
  Is there some facility in VIPER so that I can convert the the person &lt;br&gt; tracks descriptor in XML file in to text output. &lt;br&gt; Which contain the identity of person,frame and location.
  </summary>
  </entry>
  <entry>
  <author>
  <name>annie</name>
  <email>mittal.an...@gmail.com</email>
  </author>
  <updated>2008-07-10T14:04:12Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/710edef3a8bebe39/7acfa7856dc24c95?show_docid=7acfa7856dc24c95</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/710edef3a8bebe39/7acfa7856dc24c95?show_docid=7acfa7856dc24c95"/>
  <title type="text">AVI to mpg conversion for use in VIPER</title>
  <summary type="html" xml:space="preserve">
  Using AVI files needs them to be converted to mpg or image &lt;br&gt; sequence.Images sequence needs a lot of space to store for full &lt;br&gt; movie.So it will be a nightmare to store the image sequence of many &lt;br&gt; movies together. &lt;br&gt; When I use mpg encoder I lose some frames or the frame rate or frame &lt;br&gt; size is changed.
  </summary>
  </entry>
  <entry>
  <author>
  <name>annie</name>
  <email>mittal.an...@gmail.com</email>
  </author>
  <updated>2008-07-08T08:37:13Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/f830ec1b3450534f/b895023ee62902a0?show_docid=b895023ee62902a0</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/f830ec1b3450534f/b895023ee62902a0?show_docid=b895023ee62902a0"/>
  <title type="text">Re: Using AVI files</title>
  <summary type="html" xml:space="preserve">
  Thanks for the help David &lt;br&gt; The image sequence method works for me :-)
  </summary>
  </entry>
  <entry>
  <author>
  <name>David Mihalcik</name>
  <email>kraka...@gmail.com</email>
  </author>
  <updated>2008-07-07T19:59:05Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/f830ec1b3450534f/9cc961d9bf753c91?show_docid=9cc961d9bf753c91</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/f830ec1b3450534f/9cc961d9bf753c91?show_docid=9cc961d9bf753c91"/>
  <title type="text">Re: Using AVI files</title>
  <summary type="html" xml:space="preserve">
  Unfortunately, the Java Media Framework (the library ViPER uses to &lt;br&gt; display (some) AVIs) doesn&#39;t allow frame-accurate scrubbing of AVI &lt;br&gt; files, so they aren&#39;t fully supported within the program. &lt;br&gt; &lt;p&gt;I&#39;d recommend saving the video file in another format that ViPER can &lt;br&gt; read, such as MPEG-1 or 2, or a folder full of numbered jpegs.
  </summary>
  </entry>
  <entry>
  <author>
  <name>annie</name>
  <email>mittal.an...@gmail.com</email>
  </author>
  <updated>2008-07-07T18:46:01Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/f830ec1b3450534f/2c070b3488505665?show_docid=2c070b3488505665</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/f830ec1b3450534f/2c070b3488505665?show_docid=2c070b3488505665"/>
  <title type="text">Using AVI files</title>
  <summary type="html" xml:space="preserve">
  I want to use AVI files in VIPER for ground truth annotation of my &lt;br&gt; tracking frmaework but it only accepts mpg files. &lt;br&gt; Could anyone help me out how to use AVI files in VIPER
  </summary>
  </entry>
  <entry>
  <author>
  <name>Ahmad Iqbal</name>
  <email>fastah...@gmail.com</email>
  </author>
  <updated>2008-04-03T18:36:03Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/e6bc3b05cf8566e7/ecd17da625f48fb5?show_docid=ecd17da625f48fb5</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/e6bc3b05cf8566e7/ecd17da625f48fb5?show_docid=ecd17da625f48fb5"/>
  <title type="text">work in Viper</title>
  <summary type="html" xml:space="preserve">
  I wanted to make an automated algorithm that do tracking for me. &lt;br&gt; if there is some occlusion in it, I correct it using Viper Interface &lt;br&gt; than re-run the algorithm! &lt;br&gt; &lt;p&gt;Has anyone done such like task in Viper, any idea/guidelines. &lt;br&gt; &lt;p&gt;Thanks, &lt;br&gt; Ahmad Iqbal
  </summary>
  </entry>
  <entry>
  <author>
  <name>avastreg</name>
  <email>isac.new...@gmail.com</email>
  </author>
  <updated>2008-03-05T16:56:57Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/5a1cbe4899ad18cc/6f74b027b9ced8bc?show_docid=6f74b027b9ced8bc</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/5a1cbe4899ad18cc/6f74b027b9ced8bc?show_docid=6f74b027b9ced8bc"/>
  <title type="text">Re: Obox wrong matched/missed pixel calculation</title>
  <summary type="html" xml:space="preserve">
  You&#39;re right, very helpful! I&#39;ve never found that bug alone, too &lt;br&gt; complicated for me :) &lt;br&gt; &lt;p&gt;That&#39;s was the reason, i&#39;ve fixed it in the sourcecode. &lt;br&gt; &lt;p&gt;Thank you again!! &lt;br&gt; &lt;p&gt;Alessandro
  </summary>
  </entry>
  <entry>
  <author>
  <name>avastreg</name>
  <email>isac.new...@gmail.com</email>
  </author>
  <updated>2008-03-05T16:52:33Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/503ddcdf3b7ea713/92e5c91b6c458af7?show_docid=92e5c91b6c458af7</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/503ddcdf3b7ea713/92e5c91b6c458af7?show_docid=92e5c91b6c458af7"/>
  <title type="text">Re: problems about viper-pe point distance</title>
  <summary type="html" xml:space="preserve">
  Thank you for your suggestion and your answer, i&#39;ve fixed it in the &lt;br&gt; source code, checking values for null. &lt;br&gt; &lt;p&gt;Greetings &lt;br&gt; &lt;p&gt;Alessandro
  </summary>
  </entry>
  <entry>
  <author>
  <name>David Mihalcik</name>
  <email>kraka...@gmail.com</email>
  </author>
  <updated>2008-03-02T05:54:47Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/5a1cbe4899ad18cc/0f2c1a3d14f0bbd0?show_docid=0f2c1a3d14f0bbd0</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/5a1cbe4899ad18cc/0f2c1a3d14f0bbd0?show_docid=0f2c1a3d14f0bbd0"/>
  <title type="text">Re: Obox wrong matched/missed pixel calculation</title>
  <summary type="html" xml:space="preserve">
  Huh, pe seems to rotate Oriented boxes clockwise, instead of ccw. To &lt;br&gt; fix it, you have to change the &#39;initPoly&#39; method in the &lt;br&gt; edu.umd.cfar.lamp.viper.geomet ry.OrientedBox class, on line 198, add a &lt;br&gt; negative sign appropriately: &lt;br&gt; &lt;p&gt;	trans.concatenate(AffineTransf orm.getRotateInstance(Math.toR adians(-
  </summary>
  </entry>
  <entry>
  <author>
  <name>David Mihalcik</name>
  <email>kraka...@gmail.com</email>
  </author>
  <updated>2008-03-02T05:32:16Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/503ddcdf3b7ea713/16f08af440caaa13?show_docid=16f08af440caaa13</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/503ddcdf3b7ea713/16f08af440caaa13?show_docid=16f08af440caaa13"/>
  <title type="text">Re: problems about viper-pe point distance</title>
  <summary type="html" xml:space="preserve">
  There is a bug in the distance measure for points. It isn&#39;t &lt;br&gt; &#39;null safe&#39;, meaning it will fail instead of giving a distance if no &lt;br&gt; point is specified in either the target or candidate data on a frame. &lt;br&gt; &lt;p&gt;A work around is making sure you have a value on every frame that is &lt;br&gt; in your object, but that might not be doable. (e.g. a face landmark
  </summary>
  </entry>
  <entry>
  <author>
  <name>avastreg</name>
  <email>isac.new...@gmail.com</email>
  </author>
  <updated>2008-02-29T17:12:44Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/5a1cbe4899ad18cc/3a4e248800b5a447?show_docid=3a4e248800b5a447</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/5a1cbe4899ad18cc/3a4e248800b5a447?show_docid=3a4e248800b5a447"/>
  <title type="text">Obox wrong matched/missed pixel calculation</title>
  <summary type="html" xml:space="preserve">
  I have another question.. &lt;br&gt; &lt;p&gt;In the framewise evaluation, when we try to compare this target: &lt;br&gt; &lt;p&gt; &amp;lt;data:obox framespan=&amp;quot;135:135&amp;quot; height=&amp;quot;58&amp;quot; &lt;br&gt; rotation=&amp;quot;90&amp;quot; width=&amp;quot;86&amp;quot; x=&amp;quot;133&amp;quot; y=&amp;quot;150&amp;quot;/&amp;gt; &lt;br&gt; &lt;p&gt;with this candidate: &lt;br&gt; &lt;p&gt; &amp;lt;data:obox framespan=&amp;quot;135:135&amp;quot; height=&amp;quot;23&amp;quot; &lt;br&gt; rotation=&amp;quot;90&amp;quot; width=&amp;quot;27&amp;quot; x=&amp;quot;157&amp;quot; y=&amp;quot;123&amp;quot;/&amp;gt;
  </summary>
  </entry>
  <entry>
  <author>
  <name>avastreg</name>
  <email>isac.new...@gmail.com</email>
  </author>
  <updated>2008-02-29T12:13:19Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/503ddcdf3b7ea713/85c13353d502397f?show_docid=85c13353d502397f</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/503ddcdf3b7ea713/85c13353d502397f?show_docid=85c13353d502397f"/>
  <title type="text">problems about viper-pe point distance</title>
  <summary type="html" xml:space="preserve">
  Hi, i&#39;m new with ViPER-PE and i&#39;m working on it for my graduating &lt;br&gt; thesis. That&#39;s a very good work, but i&#39;ve got some problems on point &lt;br&gt; type evaluation. &lt;br&gt; &lt;p&gt;I get a NullPointerException at &lt;br&gt; viper.descriptors.attributes.A ttribute_point &lt;br&gt; $EuclideanDistance.helpGetDist ance(Unknown Source) &lt;br&gt; at viper.comparison.distances.Dis tances.. etc
  </summary>
  </entry>
  <entry>
  <author>
  <name>Miki</name>
  <email>michele...@libero.it</email>
  </author>
  <updated>2007-10-29T11:57:37Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/fd1334268de60f12/8cd0c232515fab38?show_docid=8cd0c232515fab38</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/fd1334268de60f12/8cd0c232515fab38?show_docid=8cd0c232515fab38"/>
  <title type="text">Re: Suggestions on Viper-PE</title>
  <summary type="html" xml:space="preserve">
  Hello, &lt;br&gt; how is defined &amp;quot;fragmentation&amp;quot; in framewise evaluation??? &lt;br&gt; Thank you very much..
  </summary>
  </entry>
  <entry>
  <author>
  <name>Miki</name>
  <email>michele...@libero.it</email>
  </author>
  <updated>2007-10-23T11:20:12Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/fd1334268de60f12/58eaff2119739eae?show_docid=58eaff2119739eae</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/fd1334268de60f12/58eaff2119739eae?show_docid=58eaff2119739eae"/>
  <title type="text">Re: Suggestions on Viper-PE</title>
  <summary type="html" xml:space="preserve">
  Thank you very much, you&#39;re very helpful to me. &lt;br&gt; But I don&#39;t understand one thing, yet. &lt;br&gt; This is the CONFIG file: &lt;br&gt; &amp;lt;descriptor name=&amp;quot;Descriptor_1&amp;quot; type=&amp;quot;OBJECT&amp;quot;&amp;gt; &lt;br&gt; &amp;lt;attribute dynamic=&amp;quot;true&amp;quot; name=&amp;quot;Attrib_1&amp;quot; type=&amp;quot;http:// &lt;br&gt; lamp.cfar.umd.edu/viperdata#el lipse&amp;quot;/&amp;gt; &lt;br&gt; &amp;lt;attribute dynamic=&amp;quot;true&amp;quot; name=&amp;quot;Attrib_2&amp;quot; type=&amp;quot;http://
  </summary>
  </entry>
  <entry>
  <author>
  <name>David Mihalcik</name>
  <email>kraka...@gmail.com</email>
  </author>
  <updated>2007-10-20T13:31:39Z</updated>
  <id>http://groups.google.com/group/viper-toolkit/browse_thread/thread/fd1334268de60f12/505f1fea804a1bf7?show_docid=505f1fea804a1bf7</id>
  <link href="http://groups.google.com/group/viper-toolkit/browse_thread/thread/fd1334268de60f12/505f1fea804a1bf7?show_docid=505f1fea804a1bf7"/>
  <title type="text">Re: Suggestions on Viper-PE</title>
  <summary type="html" xml:space="preserve">
  1) Filters can be specified to work on: the input or output (input &lt;br&gt; filters make it as if the filtered objects are not in the file. output &lt;br&gt; filters work on the matched objects; if either element in a match is &lt;br&gt; filtered, both are filtered. This is useful for, say, ignoring scene &lt;br&gt; text if you are evaluating an algorithm designed for overlay text.)
  </summary>
  </entry>
</feed>
