<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic IPP Equivalent of MATLAB's im2col in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Equivalent-of-MATLAB-s-im2col/m-p/997265#M22984</link>
    <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;I want to apply an operation on a sliding window over an image.&lt;BR /&gt;
	Yet it is Spatially Variant hence cannot be executed ad a convolution.&lt;/P&gt;

&lt;P&gt;My question is, does Intel IPP has something equivalent of MATLAB &lt;A href="http://www.mathworks.com/help/images/ref/im2col.html"&gt;im2col&lt;/A&gt; function (And of course the opposite way &lt;A href="http://www.mathworks.com/help/images/ref/col2im.html"&gt;col2im&lt;/A&gt;)?&lt;/P&gt;

&lt;P&gt;Those are needed in order to Vectorize the operation and keep the highest performance.&lt;/P&gt;

&lt;P&gt;Unless there are better more optimized way to engage vectorization when dealing with Spatially Variant filters.&lt;/P&gt;

&lt;P&gt;Thank You.&lt;/P&gt;</description>
    <pubDate>Mon, 21 Apr 2014 12:33:19 GMT</pubDate>
    <dc:creator>Royi</dc:creator>
    <dc:date>2014-04-21T12:33:19Z</dc:date>
    <item>
      <title>IPP Equivalent of MATLAB's im2col</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Equivalent-of-MATLAB-s-im2col/m-p/997265#M22984</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;I want to apply an operation on a sliding window over an image.&lt;BR /&gt;
	Yet it is Spatially Variant hence cannot be executed ad a convolution.&lt;/P&gt;

&lt;P&gt;My question is, does Intel IPP has something equivalent of MATLAB &lt;A href="http://www.mathworks.com/help/images/ref/im2col.html"&gt;im2col&lt;/A&gt; function (And of course the opposite way &lt;A href="http://www.mathworks.com/help/images/ref/col2im.html"&gt;col2im&lt;/A&gt;)?&lt;/P&gt;

&lt;P&gt;Those are needed in order to Vectorize the operation and keep the highest performance.&lt;/P&gt;

&lt;P&gt;Unless there are better more optimized way to engage vectorization when dealing with Spatially Variant filters.&lt;/P&gt;

&lt;P&gt;Thank You.&lt;/P&gt;</description>
      <pubDate>Mon, 21 Apr 2014 12:33:19 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Equivalent-of-MATLAB-s-im2col/m-p/997265#M22984</guid>
      <dc:creator>Royi</dc:creator>
      <dc:date>2014-04-21T12:33:19Z</dc:date>
    </item>
    <item>
      <title>hi Royi,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Equivalent-of-MATLAB-s-im2col/m-p/997266#M22985</link>
      <description>&lt;P&gt;hi Royi,&lt;/P&gt;

&lt;P&gt;it is possible, but I think will be too expensive, as requires several steps:&lt;/P&gt;

&lt;P&gt;1) transpose an image - so columns become rows&lt;/P&gt;

&lt;P&gt;2) create additional IppiSize structure with your sliding window width and height&lt;/P&gt;

&lt;P&gt;3) define the new image (sliding window) step that is == your window width*sizeof(your data type)&lt;/P&gt;

&lt;P&gt;4) one limitation of such approach - your image should be dense (step == width*sizeof(data type) or width*height of window&amp;nbsp;&amp;lt; image width&lt;/P&gt;

&lt;P&gt;for col2im you can use the same approach - transpose, etc. but I think that simple C-loop will be more efficient and doesn't have any limitations.&lt;/P&gt;

&lt;P&gt;regards, Igor&lt;/P&gt;</description>
      <pubDate>Wed, 23 Apr 2014 14:45:55 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Equivalent-of-MATLAB-s-im2col/m-p/997266#M22985</guid>
      <dc:creator>Igor_A_Intel</dc:creator>
      <dc:date>2014-04-23T14:45:55Z</dc:date>
    </item>
  </channel>
</rss>

