Regarding Standard Oxford Format for vlfeat sift
- by Karl
One of my upper classmen has gave me a data set for experimenting with vlfeat's SIFT, however, her extracted SIFT data for the frame part contains 5 dimensions. Recall from vl_sift function:
[F,D] = VL_SIFT(I)
Each column of D is the descriptor of the corresponding frame in F. F normally contains 4 dimensions which consists of x-coordinate, y-coordinate, scale, and orientation.
So I asked her what is this 5th dimension, and she pointed me to search for "standard oxford format" for sift feature.
The thing is I tried to search around regarding this standard oxford format and sift feature, but I got no luck in finding it at all. If somebody knows regarding this, could you please point me to the right direction?