博客

使用open3d来显示点云

单帧显示1:

from open3d.open3d.geometry import PointCloud
from open3d.open3d.utility import Vector3dVector
from open3d.open3d.visualization import draw_geometries

def visualize(pointcloud):
    point_cloud = PointCloud()
    point_cloud.points = Vector3dVector(pointcloud[:,0:3].reshape(-1,3))
    draw_geometries([point_cloud])

单帧显示2:

import open3d as o3d
import os

pcds = os.listdir("/home/user/Documents/reconstruction/0715_file/0806_reconstruction/")
# pcds = [i for i in pcds if i.split(".")[-1] == "pcd"]
pcds.sort()
for pcd in pcds:
    print(pcd)
    source = o3d.io.read_point_cloud("/home/user/Documents/reconstruction/0715_file/0806_reconstruction/" + pcd)  # source 为需要配准的点云

    vis = o3d.visualization.Visualizer()
    vis.create_window()
    # 将两个点云放入visualizer
    vis.add_geometry(source)
    # vis.add_geometry(target)
    vis.get_render_option().point_size = 2
    # 让visualizer渲染点云
    vis.update_geometry()
    vis.poll_events()
    vis.update_renderer()

    vis.run()

连续帧显示:

import os
import numpy as np
import open3d as o3d

files = os.listdir("0806_reconstruction/")

vis = o3d.visualization.Visualizer()
vis.create_window()
pointcloud = o3d.geometry.PointCloud()
to_reset = True
vis.add_geometry(pointcloud)
for f in files:
    pcd = o3d.io.read_point_cloud("0806_reconstruction/" + f)   #此处读取的pcd文件,也可读取其他格式的
    pcd = np.asarray(pcd.points).reshape((-1, 3))
    pointcloud.points = o3d.utility.Vector3dVector(pcd)  # 如果使用numpy数组可省略上两行
    vis.update_geometry()
    if to_reset:
        vis.reset_view_point(True)
        to_reset = False
    vis.poll_events()
    vis.update_renderer()

错误情况 1

在有些情况下,open3d会与torch或其他库冲突,只需要将

import open3d

放到第一行就可以了.

错误情况 2

如果出现open3d的错误,可以将open3d的版本换成0.8.0. 其他版本貌似会出现函数找不到的情况