首页>
外国专利>
LEARNING METHOD AND LEARNING DEVICE FOR OBJECT DETECTOR BASED ON RECONFIGURABLE NETWORK FOR OPTIMIZING ACCORDING TO CUSTOMERS' REQUIREMENTS SUCH AS KEY PERFORMANCE INDEX USING TARGET OBJECT ESTIMATING NETWORK AND TARGET OBJECT MERGING NETWORK AND TESTING METHOD AND TESTING DEVICE USING THE SAME
LEARNING METHOD AND LEARNING DEVICE FOR OBJECT DETECTOR BASED ON RECONFIGURABLE NETWORK FOR OPTIMIZING ACCORDING TO CUSTOMERS' REQUIREMENTS SUCH AS KEY PERFORMANCE INDEX USING TARGET OBJECT ESTIMATING NETWORK AND TARGET OBJECT MERGING NETWORK AND TESTING METHOD AND TESTING DEVICE USING THE SAME
A method for learning the parameters of a CNN-based object detector suitable for user requirements such as a key performance index is provided using a target object prediction network and a target object integration network. The CNN may be redesigned as the scale of an object changes due to a change in resolution or focal length according to the key performance indicators. The method includes the steps of causing the learning apparatus to output a k-th feature map by applying a convolution operation to a k-th processed image corresponding to a (k-1) target region on the image; And a step of causing the object integration network to integrate first to nth object detection information output from the FC layer, and backpropagating the loss generated by referring to the integrated object detection information and the corresponding GT. . The method has improved accuracy of a 2D bounding box, and can be usefully performed for multiple cameras, surround view monitoring, and the like.
展开▼
