<p>In <a href="https://github.com/mlpack/mlpack/pull/696#discussion_r67169108">src/mlpack/methods/edge_boxes/feature_extraction_impl.hpp</a>:</p>
<pre style='color:#555'>&gt; +      double const y_cross_x2 = y_diff * x_diff_2;
&gt; +      double const y_cross_x = y_diff * x_diff;
&gt; +      dst(row, col) = y2_cross_x2 * src(y, x) +
&gt; +              y2_cross_x * src(y, x + 1) +
&gt; +              y_cross_x2 * src(y + 1, x) +
&gt; +              y_cross_x * src(y + 1, x + 1);
&gt; +    }
&gt; +  }
&gt; +
&gt; +  return dst;
&gt; +}
&gt; +
&gt; +template&lt;typename MatType, typename CubeType&gt;
&gt; +CubeType StructuredForests&lt;MatType, CubeType&gt;::
&gt; +sepFilter2D(CubeType &amp;InOutImage, arma::vec &amp;kernel, size_t radius)
&gt; +{
</pre>
<p>Do you think it's reasonable to use the NaiveConvolution function from the mlpack code?</p>

<p><code>NaiveConvolution::Convolution(input, kernel, output)</code></p>

<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">&mdash;<br />You are receiving this because you are subscribed to this thread.<br />Reply to this email directly, <a href="https://github.com/mlpack/mlpack/pull/696/files/b38b688edfc4c2ee8bf461240f8384a70575d3a4#r67169108">view it on GitHub</a>, or <a href="https://github.com/notifications/unsubscribe/AJ4bFOzyV7HUYJdhYkAdsuXm0nU-bqm-ks5qMAilgaJpZM4I1lnr">mute the thread</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFJzQ6_ufvG5hJLykkokAOdt84AmLks5qMAilgaJpZM4I1lnr.gif" width="1" /></p>
<div itemscope itemtype="http://schema.org/EmailMessage">
<div itemprop="action" itemscope itemtype="http://schema.org/ViewAction">
  <link itemprop="url" href="https://github.com/mlpack/mlpack/pull/696/files/b38b688edfc4c2ee8bf461240f8384a70575d3a4#r67169108"></link>
  <meta itemprop="name" content="View Pull Request"></meta>
</div>
<meta itemprop="description" content="View this Pull Request on GitHub"></meta>
</div>