CNN LEARNING METHOD AND LEARNING DEVICE FOR OBJECT DETECTOR BASED ON CNN ADAPTABLE TO CUSTOMERS' REQUIREMENTS SUCH AS KEY PERFORMANCE INDEX USING TARGET OBJECT MERGING NETWORK AND TARGET REGION ESTIMATING NETWORK AND TESTING METHOD AND TESTING DEVICE USING THE SAME TO BE USED FOR MULTI-CAMERA OR SURROUND VIEW MONITORING

摘要

A method for learning the parameters of a CNN-based object detector suitable for user requirements such as key performance indicators is provided using a target object integration network and a target area prediction network. The CNN may be redesigned as the scale of the object changes as the resolution or focal length changes according to the key performance indicators. The method includes: (i) causing the target area prediction network to find the k-th prediction target area, and (ii) causing the RPN to correspond to objects on the (k_1) to (k_n) processed images. The (k_1) to (k_n) object proposals are generated, (iii) the target object integration network allows the object proposals to be integrated, and the (k_1) to (k_n) output from the FC layer. And incorporating object detection information. The above method can improve the accuracy of the 2D bounding box, and can be usefully performed for multiple cameras, surround view monitoring, and the like.