<?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 Take a look at ipps.h header in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918450#M15384</link>
    <description>Take a look at &lt;STRONG&gt;ipps.h&lt;/STRONG&gt; header files for a complete set of different IPP functions to do invert and the following operations should be supported:

...
//    DivC(v,X,Y)  :    Y&lt;N&gt; = X&lt;N&gt; / v
//    DivC(v,X)    :    X&lt;N&gt; = X&lt;N&gt; / v
//    Div(X,Y)     :    Y&lt;N&gt; = Y&lt;N&gt; / X&lt;N&gt;
//    Div(X,Y,Z)   :    Z&lt;N&gt; = Y&lt;N&gt; / X&lt;N&gt;
...&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;</description>
    <pubDate>Thu, 05 Sep 2013 13:20:48 GMT</pubDate>
    <dc:creator>SergeyKostrov</dc:creator>
    <dc:date>2013-09-05T13:20:48Z</dc:date>
    <item>
      <title>Inversion of an Image</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918448#M15382</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;
&lt;P&gt;I'm using Intel IPP 8.0&amp;nbsp;&lt;/P&gt;
&lt;P&gt;In MKL you have the option to invert a Vector / Matrix.&lt;BR /&gt;What would be the fastest way to invert an image (1 ./ A, MATLAB) using Intel IPP?&lt;/P&gt;
&lt;P&gt;I could only do it using Ln and Exp which is too slow.&lt;/P&gt;
&lt;P&gt;Could you add a native Power and Inversion commands to IPP?&lt;/P&gt;
&lt;P&gt;Thank You.&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 04 Sep 2013 14:28:58 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918448#M15382</guid>
      <dc:creator>Royi</dc:creator>
      <dc:date>2013-09-04T14:28:58Z</dc:date>
    </item>
    <item>
      <title>You can use ippsDivCRev</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918449#M15383</link>
      <description>&lt;P&gt;You can use ippsDivCRev function with a loop by rows (16u &amp;amp; 32f are supported only). For 8u the most easiest way is to create an image with the same size then use ippiSet to fill it with all ones and then use ippiDiv - it will several times faster than exp/ln approach.&lt;/P&gt;
&lt;P&gt;regards, Igor&lt;/P&gt;</description>
      <pubDate>Thu, 05 Sep 2013 07:45:01 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918449#M15383</guid>
      <dc:creator>Igor_A_Intel</dc:creator>
      <dc:date>2013-09-05T07:45:01Z</dc:date>
    </item>
    <item>
      <title>Take a look at ipps.h header</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918450#M15384</link>
      <description>Take a look at &lt;STRONG&gt;ipps.h&lt;/STRONG&gt; header files for a complete set of different IPP functions to do invert and the following operations should be supported:

...
//    DivC(v,X,Y)  :    Y&lt;N&gt; = X&lt;N&gt; / v
//    DivC(v,X)    :    X&lt;N&gt; = X&lt;N&gt; / v
//    Div(X,Y)     :    Y&lt;N&gt; = Y&lt;N&gt; / X&lt;N&gt;
//    Div(X,Y,Z)   :    Z&lt;N&gt; = Y&lt;N&gt; / X&lt;N&gt;
...&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;&lt;/N&gt;</description>
      <pubDate>Thu, 05 Sep 2013 13:20:48 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918450#M15384</guid>
      <dc:creator>SergeyKostrov</dc:creator>
      <dc:date>2013-09-05T13:20:48Z</dc:date>
    </item>
    <item>
      <title>@Seregey,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918451#M15385</link>
      <description>&lt;P&gt;@Seregey,&lt;/P&gt;
&lt;P&gt;How come none / some of them are written in the official documentation?&lt;/P&gt;
&lt;P&gt;Anyhow, none of them is a native "Image Inversion" and demand an image of "Ones" to be pre set.&lt;/P&gt;</description>
      <pubDate>Thu, 05 Sep 2013 13:47:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Inversion-of-an-Image/m-p/918451#M15385</guid>
      <dc:creator>Royi</dc:creator>
      <dc:date>2013-09-05T13:47:00Z</dc:date>
    </item>
  </channel>
</rss>

