DescriptorMatcher OpenCV train()

Question

OpenCV 的文档提到了 DescriptorMatcher 中的函数 "train()"。

"虚空 cv::cuda::DescriptorMatcher::train ( )
纯虚拟

训练描述符匹配器。

训练描述符匹配器（例如，flann 索引）。在所有要匹配的方法中，方法 train() 在匹配之前每次都是运行。"(docs)

那里就是这么说的。有人知道它有效吗？特别是 DescriptorMatcher 需要自我训练的东西。一些 OOP 语言的简短示例会很棒。

这是文档的 Link：

http://docs.opencv.org/master/dd/dc5/classcv_1_1cuda_1_1DescriptorMatcher.html#ab220b434f827962455f430a12c65c074

提前致谢

Answer 1

你可以看到匹配器代码here

Trains a descriptor matcher (for example, the flann index). In all methods to match, the method train() is run every time before matching.

是的，从代码可以看出，在匹配函数中调用了train()

void DescriptorMatcher::knnMatch( InputArray queryDescriptors, std::vector<std::vector<DMatch> >& matches, int knn,
                                  InputArrayOfArrays masks, bool compactResult )
{
    if( empty() || queryDescriptors.empty() )
        return;

    CV_Assert( knn > 0 );

    checkMasks( masks, queryDescriptors.size().height );

    train();
    knnMatchImpl( queryDescriptors, matches, knn, masks, compactResult );
}

void DescriptorMatcher::radiusMatch( InputArray queryDescriptors, std::vector<std::vector<DMatch> >& matches, float maxDistance,
                                     InputArrayOfArrays masks, bool compactResult )
{
    matches.clear();
    if( empty() || queryDescriptors.empty() )
        return;

    CV_Assert( maxDistance > std::numeric_limits<float>::epsilon() );

    checkMasks( masks, queryDescriptors.size().height );

    train();
    radiusMatchImpl( queryDescriptors, matches, maxDistance, masks, compactResult );
}

当你调用 match() 时，它实际上会调用 knnMatch 和 knn = 1

void DescriptorMatcher::match( InputArray queryDescriptors, std::vector<DMatch>& matches, InputArrayOfArrays masks )
{
    std::vector<std::vector<DMatch> > knnMatches;
    knnMatch( queryDescriptors, knnMatches, 1, masks, true /*compactResult*/ );
    convertMatches( knnMatches, matches );
}

train() 的基本实现什么都不做：

void DescriptorMatcher::train()
{}

仅FlannBasedMatcher重载train():

void FlannBasedMatcher::train()
{
    if( !flannIndex || mergedDescriptors.size() < addedDescCount )
    {
        // FIXIT: Workaround for 'utrainDescCollection' issue (PR #2142)
        if (!utrainDescCollection.empty())
        {
            CV_Assert(trainDescCollection.size() == 0);
            for (size_t i = 0; i < utrainDescCollection.size(); ++i)
                trainDescCollection.push_back(utrainDescCollection[i].getMat(ACCESS_READ));
        }
        mergedDescriptors.set( trainDescCollection );
        flannIndex = makePtr<flann::Index>( mergedDescriptors.getDescriptors(), *indexParams );
    }
}

关于如何使用FlannBasedMatcher的例子你可以参考OpenCV doc example

可以参考这个answer to know what is done in the training phase. In short, you're building the index for the matcher. You can find source code here

DescriptorMatcher OpenCV train()

DescriptorMatcher OpenCV train()

opencv

matcher

descriptor

training-data