W1995S的博客

https://blog.youkuaiyun.com/c20081052/article/details/89358658https://www.jianshu.com/p/0ef56b59b9achttps://blog.youkuaiyun.com/weixin_38715903/article/details/98039181

https://mp.weixin.qq.com/s/ZwfYJ0GpEcgKo-lZCIht8A

import numpy as npimport mayavi.mlabimport pickleimport timedef plot3Dbox(corners): for i in range(corners.shape[0]): corner = corners[i] idx = np.array([0, 1, 2, 3, 0, 4, 5, 6, 7, 4, 5, 1, 2, 6, 7, 3]) x = corner[0, idx]

Car AP_R40@0.70, 0.50, 0.50:bbox AP:95.5675, 92.1874, 91.3088bev AP:95.6500, 94.7010, 93.99183d AP:95.6279, 94.5680, 93.6853aos AP:95.54, 91.98, 90.94Pedestrian AP@0.50, 0.50, 0.50:bbox AP:65.0374, 61.3875, 57.8241bev AP:60.1475, 54.9657, 51.17

数据集从Kitti对象检测数据集下载数据（calib，image_2，label_2，velodyne），并将其放在kitti / object的数据文件夹中文件夹结构如下：kitti object testing calib image_2 label_2 velodyne training calib image_2

import syssys.getdefaultencoding()import pickleimport numpy as npnp.set_printoptions(threshold=1000000000000000)path = '/kitti_infos_test.pkl'file = open(path,'rb')inf = pickle.load(file,encoding='iso-8859-1') #读取pkl文件的内容print(inf)#fr.close(

import numpy as npfrom PIL import Imageimport matplotlib.pyplot as plt # 点云读取pointcloud = np.fromfile(str("000009.bin"), dtype=np.float32, count=-1).reshape([-1, 4])# 设置鸟瞰图范围side_range = (-40, 40) # 左右距离fwd_range = (0, 70.4) # 后前距离 x_points = p

import numpy as npimport mayavi.mlab pointcloud = np.fromfile(str("000010.bin"), dtype=np.float32, count=-1).reshape([-1,4]) print(pointcloud.shape)x = pointcloud[:, 0] # x position of pointy = pointcloud[:, 1] # y position of pointz = pointclo

Kitti数据集下载地址：http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=2d1.KITTI数据集概述KITTI数据集由德国卡尔斯鲁厄理工学院和丰田美国技术研究院联合创办，是目前国际上非常流行的自动驾驶场景下的计算机视觉算法评测数据集。该数据集用于评测立体图像(stereo)，光流(optical flow)，视觉测距(visual odometry)，3D物体检测(object detection)和3D跟踪(tr