LEARNING METHOD AND LEARNING DEVICE FOR ADJUSTING PARAMETER OF CNN BY USING MULTI-SCALE FEATURE MAP, AND TESTING METHOD AND TESTING DEVICE USING THE SAME

摘要

To provide a learning method for supporting acquisition of a bounding box corresponding to an object on a learning image from multi-scaled feature maps by using a CNN.SOLUTION: The learning method includes steps of: allowing an N-way RPN to apply certain operations to at least two specific feature maps to acquire a predetermined number of proposal boxes; allowing an N-way pooling layer to generate multiple pooled feature maps by applying pooling operations to respective areas corresponding to a predetermined number of proposal box areas on the at least two specific feature maps; and allowing an FC layer to acquire information on pixel data of the bounding box, and allowing a loss layer to acquire first comparative data, thereby adjusting at least one of parameters of the CNN by using the first comparative data during a backpropagation process.SELECTED DRAWING: Figure 3