CNN METHOD FOR TRACKING OBJECT BY USING CONVOLUTIONAL NEURAL NETWORK INCLUDING TRACKING NETWORK AND COMPUTING DEVICE USING THE SAME

摘要

The present invention relates to a method for tracking at least one object by using a convolutional neural network (CNN) including a tracking network, wherein the method comprises the steps of: (a) generating information on at least one proposal box corresponding to an object; (b) performing a process of outputting at least one position sensitive score map; (c) calculating at least one regression delta and at least one class score corresponding to the proposal box; and (d) generating at least one current bounding box corresponding to the object, and adjusting the current bounding box. According to the present invention, an object can be tracked using a CNN including a tracking network.