LEARNING METHOD AND LEARNING DEVICE FOR INTEGRATING IMAGE ACQUIRED BY CAMERA AND POINT-CLOUD MAP ACQUIRED BY RADAR OR LIDAR CORRESPONDING TO IMAGE AT EACH OF CONVOLUTION STAGES IN NEURAL NETWORK AND TESTING METHOD AND TESTING DEVICE USING THE SAME

摘要

The present invention provides a method for integrating an image generated by a camera and a corresponding point cloud map generated by a radar or lidar for each convolutional stage of a neural network, (a) a computing device, at least one cause the initial computation layer to integrate at least one original image generated by the camera with a corresponding at least one original point cloud map generated by the radar or the lidar, (i) the original point cloud At least one first fused feature map by adding depth information included in the map to the original image, and (ii) at least one second feature map by adding color information included in the original image to the original point cloud map. 1 causing a fused point cloud map to be generated; (b) cause the computing device, at least one transform layer, to generate a (1_1) first intermediate feature map by applying at least one first transform operation to the first fused feature map; generating a (1_2)th intermediate feature map by applying at least one second transform operation to the first fused point cloud map; and (c) the computing device causes at least one integration layer to generate a second fused feature map by integrating the (1_1) first intermediate feature map and the (1_2) intermediate feature map; 2 A learning method and a learning apparatus, and a test method and test apparatus using the same; and generating a second fused point cloud map by applying at least one mapping operation to the 2 fused feature map. will be.