Xueting_B的博客

代码运行环境：windows10open3d版本：0.12.0import cv2import numpy as npimport os import math import open3d as o3dbasic_path = "D:/dataset/kitti_depth/data_depth_selection/depth_selection/val_selection_cropped"image_path = "/image/2011_09_26_drive_0002_sync_i

参考连接：https://blog.csdn.net/qq_44830040/article/details/106049992?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-2.channel_param&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-2

参考连接：https://blog.csdn.net/guyuezunting/article/details/106909751计算一个4维的输入与对应的filter即卷积核的二维卷积计算tf.nn.conv2d( input, filters, strides, padding, data_format='NHWC', dilations=None, name=None)在此工程中outputs = tf.nn.conv2d(inputs, kernel,

标题：VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection作者：Yin Zhou, Oncel Tuzel来源：CVPR 2018摘要准确检测 3D 点云中的物体是许多应用中的核心问题，例如自主导航，管家机器人和 AR/VR。为了将高度稀疏的LiDAR点云与区域提议网络（RPN）连接起来，大多数现...