Real-Time Automatic Selection of the Best Shot on Object in 4K Video Stream Based on Tracking Methods in Virtual Cropped Views

Dávid Cymbalák, Ondrej Kainz, František Jakab
2015 International Journal of Computer and Electrical Engineering  
In this paper, a novel approach for real-time automatic object tracking from multiple cropped views using 4K video capturing device is presented. System itself is based on TLD object tracking algorithm. As a first ultra-high definition video and selected object tracking method is analyzed. Subsections of 4K video output from capturing device are utilized to create new, so called "virtual copped views". The developed model was rudimentary implemented and tested in the laboratory conditions and it is expected to be used in the real conditions.
doi:10.17706/ijcee.2015.7.4.275-282 fatcat:ehte4hun2vdgrcnhjwfrega5bi