Databases
StackOverflow2013
Postsdbo
POUsing OpenCV descriptor matches with findFundamentalMat
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POUsing OpenCV descriptor matches with findFundamentalMat
    primarykey
    data
    text
    <p>I posted earlier with a problem regarding the same program but received no answers. I've since corrected the issue I was experiencing at that point, only to face a new problem.</p> <p>Basically I am auto correcting stereo image pairs for rotation and translation using an uncalibrated approach. I use feature detection algorithms such as SURF to find points in two images, a left and right stereo image pair, and then using SURF again I match the points between the two images. I then need to use these matched points to find the fundamental matrix which I can use to correct the images.</p> <p>My issue is this. My matching points are stored in a single vector of descriptor matches, which is then filtered for outliers. findFundamentalMat takes as input two separate arrays of matching points. I don't know how to convert from my vector to my two separate arrays.</p> <pre><code>cout &lt;&lt; "&lt; Matching descriptors..." &lt;&lt; endl; vector&lt;DMatch&gt; filteredMatches; crossCheckMatching( descriptorMatcher, descriptors1, descriptors2, filteredMatches, 1 ); cout &lt;&lt; filteredMatches.size() &lt;&lt; " matches" &lt;&lt; endl &lt;&lt; "&gt;" &lt;&lt; endl; </code></pre> <blockquote> <p>The vector is created.</p> </blockquote> <pre><code>void crossCheckMatching( Ptr&lt;DescriptorMatcher&gt;&amp; descriptorMatcher, const Mat&amp; descriptors1, const Mat&amp; descriptors2, vector&lt;DMatch&gt;&amp; filteredMatches12, int knn=1 ) { filteredMatches12.clear(); vector&lt;vector&lt;DMatch&gt; &gt; matches12, matches21; descriptorMatcher-&gt;knnMatch( descriptors1, descriptors2, matches12, knn ); descriptorMatcher-&gt;knnMatch( descriptors2, descriptors1, matches21, knn ); for( size_t m = 0; m &lt; matches12.size(); m++ ) { bool findCrossCheck = false; for( size_t fk = 0; fk &lt; matches12[m].size(); fk++ ) { DMatch forward = matches12[m][fk]; for( size_t bk = 0; bk &lt; matches21[forward.trainIdx].size(); bk++ ) { DMatch backward = matches21[forward.trainIdx][bk]; if( backward.trainIdx == forward.queryIdx ) { filteredMatches12.push_back(forward); findCrossCheck = true; break; } } if( findCrossCheck ) break; } } } </code></pre> <blockquote> <p>The matches are cross checked and stored within filteredMatches.</p> </blockquote> <pre><code>cout &lt;&lt; "&lt; Computing homography (RANSAC)..." &lt;&lt; endl; vector&lt;Point2f&gt; points1; KeyPoint::convert(keypoints1, points1, queryIdxs); vector&lt;Point2f&gt; points2; KeyPoint::convert(keypoints2, points2, trainIdxs); H12 = findHomography( Mat(points1), Mat(points2), CV_RANSAC, ransacReprojThreshold ); cout &lt;&lt; "&gt;" &lt;&lt; endl; </code></pre> <blockquote> <p>The homography is found based on a threshold which is set at run time in the command prompt.</p> </blockquote> <pre><code>//Mat drawImg; if( !H12.empty() ) // filter outliers { vector&lt;char&gt; matchesMask( filteredMatches.size(), 0 ); vector&lt;Point2f&gt; points1; KeyPoint::convert(keypoints1, points1, queryIdxs); vector&lt;Point2f&gt; points2; KeyPoint::convert(keypoints2, points2, trainIdxs); Mat points1t; perspectiveTransform(Mat(points1), points1t, H12); for( size_t i1 = 0; i1 &lt; points1.size(); i1++ ) { if( norm(points2[i1] - points1t.at&lt;Point2f&gt;((int)i1,0)) &lt; 4 ) // inlier matchesMask[i1] = 1; } /* draw inliers drawMatches( leftImg, keypoints1, rightImg, keypoints2, filteredMatches, drawImg, CV_RGB(0, 255, 0), CV_RGB(0, 0, 255), matchesMask, 2 ); */ } </code></pre> <blockquote> <p>The matches are further filtered to remove outliers.</p> </blockquote> <p>...and then what? How do I split what's left into two Mat's of matching points to use in findFundamentalMat?</p> <p>EDIT</p> <p>I have now used my mask to make a finalMatches vector as such (this replaces the final filtering procedure above):</p> <pre><code>Mat drawImg; if( !H12.empty() ) // filter outliers { size_t i1; vector&lt;char&gt; matchesMask( filteredMatches.size(), 0 ); vector&lt;Point2f&gt; points1; KeyPoint::convert(keypoints1, points1, queryIdxs); vector&lt;Point2f&gt; points2; KeyPoint::convert(keypoints2, points2, trainIdxs); Mat points1t; perspectiveTransform(Mat(points1), points1t, H12); for( i1 = 0; i1 &lt; points1.size(); i1++ ) { if( norm(points2[i1] - points1t.at&lt;Point2f&gt;((int)i1,0)) &lt; 4 ) // inlier matchesMask[i1] = 1; } for( i1 = 0; i1 &lt; filteredMatches.size(); i1++ ) { if ( matchesMask[i1] == 1 ) finalMatches.push_back(filteredMatches[i1]); } namedWindow("matches", 1); // draw inliers drawMatches( leftImg, keypoints1, rightImg, keypoints2, filteredMatches, drawImg, CV_RGB(0, 255, 0), CV_RGB(0, 0, 255), matchesMask, 2 ); imshow("matches", drawImg); } </code></pre> <p>However I still do not know how to split my finalMatches DMatch vector into the Mat arrays which I need to feed into findFundamentalMat, please help!!!</p> <p>EDIT</p> <p>Working (sort of) solution:</p> <pre><code>Mat drawImg; vector&lt;Point2f&gt; finalPoints1; vector&lt;Point2f&gt; finalPoints2; if( !H12.empty() ) // filter outliers { size_t i, idx; vector&lt;char&gt; matchesMask( filteredMatches.size(), 0 ); vector&lt;Point2f&gt; points1; KeyPoint::convert(keypoints1, points1, queryIdxs); vector&lt;Point2f&gt; points2; KeyPoint::convert(keypoints2, points2, trainIdxs); Mat points1t; perspectiveTransform(Mat(points1), points1t, H12); for( i = 0; i &lt; points1.size(); i++ ) { if( norm(points2[i] - points1t.at&lt;Point2f&gt;((int)i,0)) &lt; 4 ) // inlier matchesMask[i] = 1; } for ( idx = 0; idx &lt; filteredMatches.size(); idx++) { if ( matchesMask[idx] == 1 ) { finalPoints1.push_back(keypoints1[filteredMatches[idx].queryIdx].pt); finalPoints2.push_back(keypoints2[filteredMatches[idx].trainIdx].pt); } } namedWindow("matches", 0); // draw inliers drawMatches( leftImg, keypoints1, rightImg, keypoints2, filteredMatches, drawImg, CV_RGB(0, 255, 0), CV_RGB(0, 0, 255), matchesMask, 2 ); imshow("matches", drawImg); } </code></pre> <p>And then I feed finalPoints1 and finalPoints2 into findFundamentalMat as Mat's. Now my only problem is that my output is not remotely as expected, the images are all screwed up :-/</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. UScherryduck
    1. UScherryduck
    plurals
    1. This table or related slice is empty.
    1. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POUsing OpenCV descriptor matches with findFundamentalMat
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POUsing OpenCV descriptor matches with findFundamentalMat
      1. USdev.microfunk
      1. VTFavorite
    3. VO
      singulars
      1. POUsing OpenCV descriptor matches with findFundamentalMat
      1. This table or related slice is empty.
      1. VTUpMod
    1. This table or related slice is empty.
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload