<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>GPU4Vision</title>
    <link>http://www.gpu4vision.org</link>
    <description>Cutting edge research results in the field of GPU-based machine vision</description>
	<atom:link href="http://gpu4vision.icg.tugraz.at/feed.xml" rel="self" type="application/rss+xml" />
    <image>
		<title> GPU4Vision </title>
        <url>http://gpu4vision.icg.tugraz.at/images/logos/logo_final_script.png</url>
		<link>http://www.gpu4vision.org</link>
    </image>
	
	
	
	

	<item>
       <title>  	  New Publication Online: A Variational Approach to Semiautomatic Generation of Digital Terrain Models (preprint) </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=stereo.php#pub38</link> 
       <description> This work will be presented at the 5th International Symposium on Visual Computing in Las Vegas, USA  </description>	   
       <pubDate>Mon, 02 Nov 2009 09:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20091102</guid>
    </item>	
	
	<item>
       <title>New Publication Online: Tracking as Segmentation of Spatial-Temporal Volumes by Anisotropic Weighted TV</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=segmentation.php#pub35</link> 
       <description> This work will be presented at the EMMCVPR in Bonn, Germany </description>	   
       <pubDate>Wed, 12 Aug 2009 09:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090812</guid>
    </item>		
	
	<item>
       <title>New Preprint Online: Global Solutions of Variational Models with Convex Regularization  </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=stereo.php#pub34</link> 
       <description> In this paper, we propose an algorithmic framework to compute global solutions of variational methods with convex regularity terms but which can handle quite arbitrary data terms. While the minimization of variational problems with convex data and regularity terms is straight forward (using for example gradient descent), this is no longer trivial for functionals with non convex data terms...  </description>	   
       <pubDate>Tue, 11 Aug 2009 10:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090811a</guid>
    </item>	
	
	<item>
       <title>New Preprint Online: Total Generalized Variation </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=industry.php#pub33</link> 
       <description> The novel concept of total generalized variation of a function u is introduced and some of its essential properties are proved. Differently from the bounded variation semi-norm, the new concept involves higher order derivatives of u...  </description>	   
       <pubDate>Tue, 11 Aug 2009 09:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090811</guid>
    </item>		
	
	<item>
       <title>New Publication Online: Anisotropic Huber-L1 Optical Flow</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=motion.php#pub32</link> 
       <description> This work will be presented at the British Machine Vision Conference 2009 in London, UK </description>	   
       <pubDate>Wed, 05 Aug 2009 09:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090805</guid>
    </item>	
	
    <item>
       <title>New Publication Online: An Algorithm for Minimizing the Mumford-Shah Functional  </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=segmentation.php#pub28</link> 
       <description> This work will be presented at the International Conference on Computer Vision 2009 in Kyoto, Japan </description>	   
       <pubDate>Thu, 16 Jul 2009 13:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090716a</guid>
    </item>	
	
    <item>
       <title>New Publication Online: Interactive Texture Segmentation using Random Forests and Total Variation </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=segmentation.php#pub27</link> 
       <description>This work will be presented at the British Machine Vision Conference 2009 in London, UK</description>	   
       <pubDate>Thu, 16 Jul 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090716</guid>
    </item>
	
	<item>
       <title>Software Update: FlowLib </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub26Bin1</link> 
       <description> Update of our FlowLib package to support CUDA 2.2. Linux x64 and Win32 binaries are included in this new package. This optical flow implementation is the successor of our famous OFLib. It ports our recently published Matlab package to the GPU.</description>	   
       <pubDate>Fri, 26 Jun 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090626</guid>
    </item>	
	
	<item>
       <title>New Software Online: FlowLib </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub26Bin1</link> 
       <description>This new optical flow implementation is the successor of our OFLib. It ports our recently published Matlab package to the GPU.</description>	   
       <pubDate>Tue, 21 Apr 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090421</guid>
    </item>
	
	<item>
       <title>New Publication Online: A Variational Model for Interactive Shape Prior Segmentation and Real-Time Tracking </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=segmentation.php#pub25</link> 
       <description>This work will be presented at the 2nd International Conference on Scale Space and Variational Methods in Computer Vision in Voss, Norway</description>	   
       <pubDate>Fri, 17 Apr 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090417</guid>
    </item>

	<item>
       <title>New Publication Online: Semi Automatic Segmentation of Articular Cartilage using Variational Methods</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=processing.php#pub24</link> 
       <description>This master's thesis describes an interactive segmentation framework for the semi automatic segmentation of articular cartilage.</description>	   
       <pubDate>Fri, 10 Apr 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090410</guid>
    </item>
	
    <item>
       <title>New Software Online: Matlab optical flow package</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub3Bin2</link> 
       <description>With this new software package, you can directly compute duality-based TV-L1 optical flow using Matlab. </description>	   
       <pubDate>Wed, 18 Mar 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090318</guid>
    </item>	
	
    <item>
       <title>New Publication Online: A Convex Relaxation Approach for Computing Minimal Partitions</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=segmentation.php#pub23</link> 
       <description>This work will be presented at the IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2009 in Miami, Florida, USA</description>	   
       <pubDate>Mon, 16 Mar 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090316</guid>
    </item>	
	
    <item>
       <title>New Student Project Offer</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=news.php</link> 
       <description>We're currently looking for a student willing to join our research group for a bachelor's thesis or seminar project. </description>	   
       <pubDate>Wed, 25 Feb 2009 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20090225</guid>
    </item>	
	
	<item>
       <title>NVIDIA contributes to GPU4Vision. Again.</title>
       <link>http://www.nvidia.com/object/tesla_c1060.html</link> 
       <description>After gratefully having received 2x 8800 GTX in 2007 and 2x 9800 GX2 in mid-2008, we're proud to announce that NVIDIA supports our research projects again by donating the brand new high performance computing solution Tesla C1060. This cutting edge card equipped with 4 GB of GDDR will be used to tackle large-scale problems with enormous memory consumption.
</description>	   
       <pubDate>Tue, 11 Nov 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081111</guid>
    </item>
	
	<item>
       <title>New Publication Online: Duality TV-L1 Flow with Fundamental Matrix Prior</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=motion.php#pub21</link> 
       <description>This work will be presented at the International Conference Image and Vision Computing 2008 in Auckland (New Zealand) </description>	   
       <pubDate>Wed, 05 Nov 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081105</guid>
    </item>	
	
    <item>
       <title>Taking the lead...</title>
       <link>http://vision.middlebury.edu/flow/eval/results/results-e1.html</link> 
       <description>With an improved TVL1 optical flow algorithm based on our DAGM2007 paper 'A Duality Based Approach for Realtime TV-L1 Optical Flow', we have taken the lead in the Middlebury database evaluation for both angular error and end-point error! A new paper describing the improvements in detail will soon be published here, so keep an eye on our website...</description>	   
       <pubDate>Thu, 23 Oct 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081023</guid>
    </item>		
	
	<item>
       <title>New Publication Online: Continuous Energy Minimization via Repeated Binary Fusion</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=motion.php#pub1</link> 
       <description>This work has been presented at the European Conference on Computer Vision 2008 in Marseilles (France) </description>	   
       <pubDate>Tue, 21 Oct 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081021</guid>
    </item>	
	
	<item>
       <title>VMFLib online </title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub4Bin1</link> 
       <description>The Variational Methods Filter Library (VMFLib 0.1) is now online in its first version. A Windows 32bit library is provided together with a sample code showing its usage. Currently two different denoising models are supported: ROF and TV-L1. Different minimizers can be applied to 2D grayscale images. </description>	   
       <pubDate>Wed, 15 Oct 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081015</guid>
    </item>		
	
	<item>
       <title>Flowball video online</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub18Video1</link> 
       <description>Flowball is an interactive game presented at the Open Lab Night of our Insitute in October 2008. Using dense optical flow computed in realtime on a Geforce GTX 280, flowball is a demonstration of current capabilities of GPGPU for the interested public. You can also watch a low quality version on our Youtube channel gpu4vision.</description>	   
       <pubDate>Thu, 09 Oct 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081009</guid>
    </item>	
	
	<item>
       <title>Open student projects </title>
       <link>http://www.icg.tugraz.at/Opportunities</link> 
       <description>We're currently looking for motivated students, who'd like to participate in our research group. We offer different seminar projects or thesis projects, all available at our Institute's website. If you're interested, don't hesitate to contact us.</description>	   
       <pubDate>Sun, 05 Oct 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081005</guid>
    </item>		
	
	<item>
       <title>GPU4Vision contributes to Folding@Home</title>
       <link>http://fah-web.stanford.edu</link> 
       <description>GPU4Vision started to contribute to Folding@Home, a distributed computing application with the goal to understand protein folding processes. The FoH team started to provide high performance GPU-Clients, so we began to contribute GPU computation time. If you have an idle GPU and would like to contribute too, you can be part of the GPU4Vision team by using our team ID 136729.</description>	   
       <pubDate>Wed, 01 Oct 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20081001</guid>
    </item>		
	
	<item>
       <title>Videos available on Youtube</title>
       <link>http://www.youtube.com/user/gpu4vision</link> 
       <description>Some of our most descriptive videos can be viewed directly on Youtube. We will also show videos of our current work there, as well as videos which are not part of our latest publications. Our channel is called gpu4vision, we'd appreciate some nice ratings :-).</description>	   
       <pubDate>Mon, 25 Aug 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080825</guid>
    </item>
	
    <item>
       <title>New Binary Online: TVSeg 0.2</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub6Bin2</link> 
       <description>A Win32 binary of our interactive segmentation tool (TVSeg) is now available for download. Version 0.2 features a more stable energy minimization, automatic stopping, background images, test images  and new tutorials.</description>	   
       <pubDate>Thu, 14 Aug 2008 13:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080814a</guid>
    </item>	
	
	<item>
       <title>New Publication Online: TVSeg - Interactive Total Variation Based Image Segmentation</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=segmentation.php#pub6</link> 
       <description>This work will be presented at the British Machine Vision Conference 2008 in Leeds (UK)</description>	   
       <pubDate>Thu, 14 Aug 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080814</guid>
    </item>
	
	<item>
       <title>New Publication Online: Automatic Differentiation for GPU-Accelerated 2D/3D Registration</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=motion.php#pub16</link> 
       <description>This work has been presented at the International Conference on Automatic Differentiation in Bonn (Germany)</description>	   
       <pubDate>Tue, 12 Aug 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080812</guid>
    </item>
	
	<item>
       <title>New Publication Online: A Convex Formulation of Continuous Multi-Label Problems</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=stereo.php#pub14</link> 
       <description>This work will be presented at the European Conference on Computer Vision in Marseilles (France)</description>	   
       <pubDate>Mon, 11 Aug 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080811</guid>
    </item>
	
    <item>
       <title>New Library Online: OFLib 0.1</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub3Bin1</link> 
       <description>A Win32 library providing GPU powered dense realtime optical flow is available in our download section.</description>	   
       <pubDate>Mon, 04 Aug 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080804</guid>
    </item>	
	
	<item>
       <title>CUDA Templates made public on SourceForge.net</title>
       <link>https://sourceforge.net/projects/cudatemplates</link> 
       <description>CUDA Templates is a collection of C++ template classes and functions which provide a consistent interface to NVIDIA's CUDA, hiding much of the complexity of the underlying CUDA functions from the programmer. 
By making CUDA Templates publicly available on SourceForge, we want to make the development of your GPU projects faster and easier, and of course encourage you to contribute to this helpful piece of code."
</description>	   
       <pubDate>Tue, 22 Jul 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080722</guid>
    </item> 
	
	<item>
       <title>NVIDIA accepts GPU4Vision as featured project</title>
       <link>http://www.nvidia.com/object/cuda_home.html</link> 
       <description>NVIDIA presents selected academic and commercial projects which use their graphics cards on their webpage. With the linkage to www.GPU4Vision.org, NVIDIA once again promotes our hard work. </description>	   
       <pubDate>Mon, 21 Jul 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080721</guid>
    </item>	
	
	 <item>
       <title>New Binary Online: TVSeg 0.1</title>
       <link>http://gpu4vision.icg.tugraz.at/index.php?content=downloads.php#pub6Bin2</link> 
       <description>A Win32 binary of our interactive segmentation tool is now available for download.</description>	   
       <pubDate>Wed, 16 Jul 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080716</guid>
    </item> 
	
    <item>
       <title>NVIDIA contributes to our project</title>
       <link>http://www.nvidia.com/object/geforce_9800gx2.html</link> 
       <description>After providing two 8800GTX in 2007, NVIDIA again supported our project with two of their 9800GX2 cutting edge graphics adapters. We'd like to thank NVIDIA for their continuing support.</description>	   
       <pubDate>Thu, 19 Jun 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080618</guid>
    </item>	
	
    <item>
       <title>GPU4Vision launched</title>
       <link>http://www.gpu4vision.org</link> 
       <description>The GPU4Vision project founded by the Institute for Computer Graphics and Vision, Graz University of Technology has started its web appearance www.gpu4vision.org to make the latest research results publicly available.</description>	   
       <pubDate>Wed, 18 Jun 2008 12:00:00 GMT</pubDate>
	   <guid isPermaLink="false">www.gpu4vision.org_rss_item_20080619</guid>
    </item> 
   
  </channel>
</rss> 