UDC 004.932
SCALING OF GROUND TRUTH FOR STEREO TASKS
E. R. Muratov, Ph.D. (Tech.), assistant professor, RSREU, Ryazan, Russia;
orcid.org/0000-0002-1664-3954, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The article proposes an algorithm for reducing the resolution of depth maps and ground truth, estimating and comparing error metrics resulting from a reduction size of ground truth and images for various scaling algorithms. The aim of the work is to estimate the accuracy of scaling ground truth by the proposed algorithm. The need for scaling stereo dataset arises in the task of obtaining data sets with uniform parameters of a stereo system, one of which is sensor resolution, from existing different of stereo datasets. A generalized data set with the same parameters of stereo system is required for designing more accurate hybrid stereo vision algorithms with the elements of artificial intelligence.
Key words: depth maps scaling, downsampling ground truth, dense depth map, depth map accuracy, BAD error estimation metric, image scaling, scaling algorithms comparison.