baidu_40840693的博客

可视化 相机pose 或者 pose视锥

cmakelist.txtcmake_minimum_required(VERSION 3.19)set (CMAKE_CXX_STANDARD 14)project(TestVisualizer)# Find installed Open3D, this imports# Open3D_INCLUDE_DIRS# Open3D_LIBRARIES# Open3D_LIBRARY_DIRS# Open3D_C_FLAGS# Open3D_CXX_FLAGS#

代码来源：https://github.com/agarwa65/lidar_camera_calibration最核心的只是把校准函数calibrate的核心crop，去理解整个过程，其实和 结构光相机与RGB相机对齐，没有区别import numpyimport math#from tf.transformations import euler_from_matrix #原函数# axis sequences for Euler angles_NEXT_AXIS = [1, .

http://ivl.calit2.net/wiki/images/0/00/07_ProjectionAndFrustumCullingF19.pdfhttps://my.eng.utah.edu/~cs6360/Lectures/frustum.pdfModel View Projectionopengl 交互式控制8.3 - Perspective Projections — LearnWebGL

from __future__ import print_functionimport urllibimport urllib.requestimport bz2import osimport numpy as npnp.set_printoptions(suppress=True)import pclimport pcl.pcl_visualizationimport randomdef vis_pair(cloud1, cloud2, rdm=False): color.

GitHub - rpautrat/SuperPoint: Efficient neural feature detector and descriptor1.checkpoint转换pbSuperPoint/export_model.py at master · rpautrat/SuperPoint · GitHub2.freeze_graph参考1：Python freeze_graph.freeze_graph方法代碼示例Python freeze_graph.freeze_

superGlue学习_u010949023的博客-优快云博客superPoint学习_u010949023的博客-优快云博客_superpoint一种基于注意力机制特征匹配网络SuperGlue：端到端深度学习SLAM的重要里程碑_3D视觉工坊-优快云博客

cmake_minimum_required(VERSION 3.12)project(PCL_demo1) set(CMAKE_CXX_STANDARD 14) find_package(OpenCV REQUIRED)find_package(PCL REQUIRED) include_directories(${PCL_INCLUDE_DIRS})link_directories(${PCL_LIBRARY_DIRS})add_definitions(${PCL_DEFINI.

WideBaselineFeatureMatcher[1] CODE: Coherence Based Decision Boundaries for Feature Correspondence, IEEE TPAMI,2016, Lin et. al.[2] RepMatch: Robust Feature Matching and Pose for Reconstructing Modern Cities, ECCV, 2016,, Lin et. al.高斯过程 Gaussian Pr

图像分割概述图像分割概述Tomasi-Kanade 因子分解Tomasi-Kanade 因子分解观测矩阵的秩 _ Structure from Motion观测矩阵的秩 _ Structure from Motion观测矩阵 _ Structure from Motion观测矩阵 _ Structure from Motion运动恢复结构问题 Structure from Motion运动恢复结构问题 Structure from MotionStructure fr

import cv2import numpy as npimport matplotlib.pyplot as pltturbo_colormap_data = np.array( [[0, 0, 0], #[0.18995, 0.07176, 0.23217], [0.19483, 0.08339, 0.26149], [0.19956, 0.09498, 0.29024], [0.20415, 0.10652, 0.31844], [0.

tensorflow：https://github.com/rpautrat/SuperPointimport cv2import numpy as np#opencv-python==3.4.2.17#opencv-contrib-python==3.4.2.17class LocalImageDescriptor(object): def __init__(self,): octaves = 3 self.sift_pt = cv2.xfea

import osimport cv2import jsonimport numpy as npimport copy#opencv/modules/imgproc/src/geometry.cppwith open(".json", "r") as f: load_dict = json.load(f)# 只标注一个mask测试m_value = load_dict["shapes"][0]points = np.array(m_value["points"])co...

参考：https://blog.youkuaiyun.com/aihua53/article/details/112466439https://blog.youkuaiyun.com/qq_41917697/article/details/114437924Linux环境变量查看方法1：echo $PATH方法2：export -p | grep PATH我的~/.bashrc中是在当前终端输入：export CUDA_HOME=/usr/local/cuda-10.0export PATH=.

# yangninghuaimport cv2import open3dimport numpy as npimport osimport shutilimport sysimport jsonfrom PIL import Imageimport matplotlibimport matplotlib.pyplot as pltturbo_colormap_data = np.array( [[0, 0, 0], #[0.18995, 0.07176, ...

安装gcc7,默认gcc9太高了sudo vim /etc/apt/sources.listdeb http://dk.archive.ubuntu.com/ubuntu/ xenial maindeb http://dk.archive.ubuntu.com/ubuntu/ xenial universesource /etc/apt/sources.listsudo apt update sudo apt install gcc-5 g++-5 sudo apt instal

import osimport sysimport cv2import cv2 as cvimport timeimport jsonimport copyimport mathimport pathlibimport numpy as npdef affine_skew(tilt, phi, img, mask=None): ''' affine_skew(tilt, phi, img, mask=None) -> skew_img, skew_mask, .

https://blog.youkuaiyun.com/taifyang/article/details/105248483https://github.com/charlesq34/pointnet#!/usr/bin/env python# coding: utf-8import open3d as o3dimport numpy as npimport osimport matplotlib.pyplot as pltfrom mpl_toolkits.mplot3d import Ax.

# examples/Python/Advanced/non_blocking_visualization.pyimport open3d as o3dimport numpy as npimport copydef save_view_point(pcd, filename): vis = o3d.visualization.Visualizer() vis.create_window() vis.add_geometry(pcd) vis.run(...

参考:https://github.com/intel-isl/Open3D/issues/1367

pip install cmake pybind11 ninjaconda install -c conda-forge cmake pybind11 ninja ceres-solver

12*925*25的棋盘格# python3.8# -*- coding: utf-8 -*-# ---# @Software: PyCharm# @File: camera_calibration.py# @Author: ---# @Institution: BeiJing, China# @E-mail: lgdyangninghua@163.com# @Site: # @Time: 5月 14, 2021# ---# Import required modul..

hex_head2.scad$fn=100;module hex_head(diameter,height) { intersection() { union() { translate([0,0,0.801*height]) cylinder(d1=diameter,d2=0.8*diameter,h=0.2*height); cylinder(d=diameter,h=0.801*height); } cylin.

(Camera Calibration | Camera Calibration)

输出的ply无法在cloudcompare上打开我切换成pcd进行打开# python3.8# -*- coding: utf-8 -*-# ---# @Software: PyCharm# @File: vis_ply.py# @Author: ---# @Institution: BeiJing, China# @E-mail: lgdyangninghua@163.com# @Site: # @Time: 4月 26, 2021# ---import osimpor

open3d - voxel_down_sample#open3d - voxel_down_sampledef VoxelDownSample(PointCloud, voxel_size): #参考：https://www.cnblogs.com/huashanqingzhu/p/6603875.html INT_MAX = 2147483647 INT_MIN = -2147483648 output = [] if voxel_size <=.

如何入门线性代数？这里有一份Python线性代数讲义https://github.com/MacroAnalyst/Linear_Algebra_With_PythonQR分解：https://github.com/MacroAnalyst/Linear_Algebra_With_Python/blob/master/Chapter%2016%20-%20Gram-Schmidt%20Process%20and%20QR%20Decomposition.ipynbSVD分解：https.

https://scipy-cookbook.readthedocs.io/items/OLS.html

代码去适配python3,但是，多进程那边bug没解决，由于我不太需要那个，就没管了代码来自于：https://github.com/scipy/scipy-cookbook/# python3.8# -*- coding: utf-8 -*-# ---# @Software: PyCharm# @File: main_kd-tree.py# @Author: yangninghua# @Institution: ~ ,BeiJing, China# @E-mail: lgdya

import numpyimport scipy # use numpy if scipy unavailableimport scipy.linalg # use numpy if scipy unavailable## Copyright (c) 2004-2007, Andrew D. Straw. All rights reserved.## Redistribution and use in source and binary forms, with or without##...

python BA优化示例from __future__ import print_functionimport urllibimport urllib.requestimport bz2import osimport numpy as npnp.set_printoptions(suppress=True)import pclimport pcl.pcl_visualizationimport randomdef vis_pair(cloud1, cloud2, rdm=F.

https://openmvg.readthedocs.io/en/latest/software/SfM/GlobalSfM/[GlobalACSfM]is based on the paper “Global Fusion of Relative Motions for Robust, Accurate and Scalable Structure from Motion.” published at ICCV 2013.http://theia-sfm.org/sfm.html...

https://www.opensfm.org/docs/building.html在Ubuntu系统安装配置OpenSfM软件用来实现三维重建 ***https://www.jianshu.com/p/466a6679c792openSFM配置心得https://blog.youkuaiyun.com/baiyang10010/article/details/89031439OpenSFM配置细节（dockerfile使用）https://blog.youkuaiyun.com/baiyang10010/

可视化资料https://matejsladek.com/data/bachelor_thesis.pdfThree.jshttps://publik.tuwien.ac.at/files/publik_253569.pdf openSFMhttps://yangliu.life/build/misc/thesis.pdfThree.jshttps://developer.mozilla.org/en-US/docs/Web/API/WebGL_API/WebGL_model_vi...

还没写完，工作事太多了。。。#create by yangninghua#2021.4.4import osimport cv2import sysimport copyimport mathimport argparseimport numpy as npfrom matplotlib import pyplot as pltdef mkdir_os(path): if not os.path.exists(path): os.makedirs(p.

https://github.com/PacktPublishing/Mastering-OpenCV3-Second-Edition具体环境配置参考我上一篇博文：https://blog.youkuaiyun.com/baidu_40840693/article/details/115284208上一篇我们已经安装好了opencv4 vtk9.0我们在 vcpkg-e054fe2b20469c894b88be4ac466776f9f653954 分支上继续编译pcl遇到了一个问题：...

1#include <opencv2/viz.hpp>#include <iostream>using namespace cv;using namespace std;void help(){ cout << "--------------------------------------------------------------------------" << endl << "This progra

参考：https://blog.youkuaiyun.com/AIchipmunk/article/details/48132109https://blog.youkuaiyun.com/AIchipmunk/article/details/48157369https://blog.youkuaiyun.com/AIchipmunk/article/details/51232861https://blog.youkuaiyun.com/aichipmunk/article/details/52433884https://github.

Explore Structure from Motion with the SfM Module需要的东西vcpkg 版本vcpkg-2020.111. 首先vs2015安装，然后语言变成English2. powershell v7.0.3以上3. 由于我没有选择VS2015默认安装路径，error：https://github.com/microsoft/vcpkg/issues/12488修改：D:\install\vcpkg\vcpkg-2020.11\toolsrc\s.

opencv3.4.12及contrib在ubuntu1604编译https://zhuanlan.zhihu.com/p/265854664Windows下OpenCV3.4.0 SFM模块安装配置https://www.meiwen.com.cn/subject/ouqrjftx.htmltutorial_sfm_scene_reconstructionhttps://docs.opencv.org/master/d4/d18/tutorial_sfm_scene_reconstru