Computes an image descriptor using the set visual vocabulary
[bow,idx,kptDescs] = extractor.compute(img, keypoints)
Input
- img Image, for which the descriptor is computed.
- keypoints Keypoints detected in the input image. It is a
struct array that is returned by cv.FeatureDetector.detect.
Output
- bow Computed output image descriptor. A vector of the same
length as the vocabulary dimension.
- idx Indices of keypoints that belong to the cluster. A
cell array of integer vectors. This means that
idx{i} are
keypoint indices that belong to the i-th cluster (word of
vocabulary).
- kptDescs Descriptors of the image keypoints, as returned
by cv.DescriptorExtractor.compute.