RGBD-slam和DVO-slam安装及运行

已于 2022-03-16 10:24:00 修改
阅读量2.3k 收藏 9
点赞数 1
分类专栏： SLAM 文章标签： 计算机视觉
于 2022-03-14 21:30:56 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_41469272/article/details/123488687
版权

文章目录

RGBD-slam

主要参考https://github.com/felixendres/rgbdslam_v2

1 下载

（1）源码包下载

cd <ros_ws>/src
git clone https://github.com/felixendres/rgbdslam_v2.git

尖括号<>中的内容需根据自己需要自行配置的意思
（2）下载作者提供的g2o

cd </home/xxx/Download/softpackages>
git clone https://github.com/felixendres/g2o.git
cd g2o & mkdir build install & cd build
cmake .. -DCMAKE_INSTALL_PREFIX=</home/xxx/Download/softpackages>/g2o/install
make & make install

（3） 修改以下文件：

1）src/misc.cpp

801：
-	inline int round(float d)
+  inline int ROUND(float d)

877：
-      int old_rx_center = round((p.x / p.z)* fx + cx);
-      int old_ry_center = round((p.y / p.z)* fy + cy);
+      int old_rx_center = ROUND((p.x / p.z)* fx + cx);
+      int old_ry_center = ROUND((p.y / p.z)* fy + cy);

1012：
-      int old_rx_center = round((p.x / p.z)* fx + cx);
-      int old_ry_center = round((p.y / p.z)* fy + cy);
+      int old_rx_center = ROUND((p.x / p.z)* fx + cx);
+      int old_ry_center = ROUND((p.y / p.z)* fy + cy);

2）package.xml
加两行：

+  <build_depend>cv_bridge</build_depend>
+  <run_depend>cv_bridge</run_depend>

3）launch/rgbdslam.launch

6：
-<param name="config/topic_image_depth"             value="/camera/depth_registered/sw_registered/image_rect_raw"/>
+<param name="config/topic_image_depth" value="/camera/depth/image"/>

Tips:

本文将依赖的包均安装在/home/xxx/Download/softpackages下
cv_bridge使用版本：melodic分支
OpenCV：3.4.14
G2o：c++03分支

4）Cmakelist.txt
1、 关闭gpu，可选项

-set(USE_SIFT_GPU  1 CACHE BOOL "build with support for siftgpu") 
+set(USE_SIFT_GPU  0 CACHE BOOL "build with support for siftgpu")

2、添加c++11标准

+add_compile_options(-std=c++11)

3、设置cv_bridge

+# cv_bridge
+set(cv_bridge_INCLUDE_DIRS /home/xxx/Download/softpackages/vision_opencv/cv_bridge/install/include)
+set(cv_bridge_LIBRARY_DIRS /home/xxx/Download/softpackages/vision_opencv/cv_bridge/install/lib)
+include_directories(${cv_bridge_INCLUDE_DIRS})
+link_directories(${cv_bridge_LIBRARY_DIRS})

4、设置g2o

# g2o #######################
+set(G2O_DIR /home/xxx/Download/softpackages/g2o/build)
find_package(G2O REQUIRED)

5、 设置opencv

+set(OpenCV_DIR /home/xxx/Download/softpackages/opencv/build)
find_package(OpenCV REQUIRED)

6、设置链接选项

-SET(LIBS_LINK GL GLU cholmod cxsparse Qt5::Widgets Qt5::OpenGL Qt5::Concurrent ${G2O_LIBRARIES} ${GLUT_LIBRARY} ${OPENGL_LIBRARY} ${OpenCV_LIBS})
+SET(LIBS_LINK GL GLU cholmod cxsparse Qt5::Widgets Qt5::OpenGL Qt5::Concurrent ${G2O_LIBRARIES} ${GLUT_LIBRARY} ${OPENGL_LIBRARY} ${OpenCV_LIBS} cv_bridge)

5）G2o的G2OConfig.cmake文件内容：

get_filename_component(G2O_CMAKE_DIR "${CMAKE_CURRENT_LIST_FILE}" PATH) 


set(G2O_INCLUDE_DIRS "/home/xxx/workspace/openslam_nav/src/rgbdslam_v2/lib/g2o/install/include") 
set(G2O_LIBRARIES)

# Set variable for individual library and also append to G2O_LIBRARIES
macro(find_g2o_library VARNAME LIBRARYNAME)
find_library(${VARNAME}
  NAMES "g2o_${LIBRARYNAME}"
  PATHS
  /home/xxx/workspace/openslam_nav/src/rgbdslam_v2/lib/g2o/install/lib
  NO_DEFAULT_PATH
  )
#Append to the global list of all libraries
list(APPEND G2O_LIBRARIES ${${VARNAME}})
endmacro(find_g2o_library VARNAME LIBRARYNAME)

# Now populate the variables
# Find the core elements
find_g2o_library(G2O_STUFF_LIBRARY stuff)
find_g2o_library(G2O_CORE_LIBRARY core)

# Find the CLI library
find_g2o_library(G2O_CLI_LIBRARY cli)

# Find the pluggable solvers
find_g2o_library(G2O_SOLVER_CHOLMOD solver_cholmod)
find_g2o_library(G2O_SOLVER_CSPARSE solver_csparse)
find_g2o_library(G2O_SOLVER_CSPARSE_EXTENSION csparse_extension)
find_g2o_library(G2O_SOLVER_DENSE solver_dense)
find_g2o_library(G2O_SOLVER_PCG solver_pcg)
find_g2o_library(G2O_SOLVER_SLAM2D_LINEAR solver_slam2d_linear)
find_g2o_library(G2O_SOLVER_STRUCTURE_ONLY solver_structure_only)
find_g2o_library(G2O_SOLVER_EIGEN solver_eigen)

# Find the predefined types
find_g2o_library(G2O_TYPES_DATA types_data)
find_g2o_library(G2O_TYPES_ICP types_icp)
find_g2o_library(G2O_TYPES_SBA types_sba)
find_g2o_library(G2O_TYPES_SCLAM2D types_sclam2d)
find_g2o_library(G2O_TYPES_SIM3 types_sim3)
find_g2o_library(G2O_TYPES_SLAM2D types_slam2d)
find_g2o_library(G2O_TYPES_SLAM3D types_slam3d)

2 编译

(1) 编译

cd <ros_ws>
catkin_make -DCATKIN_WHITELIST_PACKAGES=rgbdslam

(2) 查看：rgbdslam链接的g2o版本

ldd rgbdslam |grep g2o
libg2o_stuff.so => /home/xxx/workspace/openslam_nav/src/dvo_slam/thirdparty/g2o/install/lib/libg2o_stuff.so (0x00007f089599c000)
……

若链接到/opt/ros/melodic下，则会报未定义的问题，如下

libg2o_stuff.so => /opt/ros/melodic/lib/libg2o_stuff.so

则有两种解决办法：
1） 将/opt/ros/melodic下的g2o的库在编译前移走
2）使用patchelf更改链接路径：

cd <ros_ws>/devel/lib/rgbdslam
patchelf --replace-needed libg2o_stuff.so /home/xxx/workspace/openslam_nav/src/rgbdslam_v2/lib/g2o/lib/libg2o_stuff.so rgbdslam

3 运行

cd <ros_ws>
source devel/setup.sh
roslaunch rgbdslam rgbdslam.launch
#kitti rgbd dataset
rosbag play rgbd_dataset_freiburg1_360.bag

参考链接：
https://github.com/felixendres/rgbdslam_v2
https://wiki.ros.org/rgbdslam_electric/evaluation

在这里插入图片描述

DVO 编译运行：

主要参考：https://github.com/songuke/dvo_slam

下载DVO源码及依赖库文件

1.下载dvo-slam

cd <ros_ws>/src
git clone https://github.com/songuke/dvo_slam.git
cd dvo_slam
mkdir thirdparty

2.下载编译Sophus

cd thirdparty
git clone https://github.com/strasdat/Sophus.git
cd Sophus
git checkout b474f05
mkdir build
cd build
cmake ..
make

3.下载编译g2o

1)下载g2o

cd thirdparty
git clone https://github.com/RainerKuemmerle/g2o.git
cd g2o
git checkout 67d5fa7

2）下载Eigen 3.2.8.

cd thirdparty/g2o/EXTERNAL
git clone https://github.com/eigenteam/eigen-git-mirror.git
git checkout 3.2.8

3）编译g2o

cd thirdparty/g2o
mkdir build install
cd build
cmake .. -DG2O_BUILD_APPS:BOOL=OFF -DG2O_BUILD_EXAMPLES:BOOL=OFF -DG2O_EIGEN3_INCLUDE=/home/xxx/workspace/openslam_nav/src/dvo_slam/thirdparty/g2o/EXTERNAL/eigen-3.2.8 -DCMAKE_INSTALL_PREFIX=/home/xxx/workspace/openslam_nav/src/dvo_slam/thirdparty/g2o/install
make&make install

4）创建G2OConfig.cmake，对应修改G2O_INCLUDE_DIRS及设置G2O_LIBRARIES，参见G2OConfig.cmake

4.修改dvo的几个package的CMakeLists.txt

（1）dvo_core的CMakeLists.txt
Path：dvo_slam/src/dvo_core/CMakeLists.txt

cmake_minimum_required(VERSION 2.8.3)

project(dvo_core)

set(Sophus_DIR ${PROJECT_SOURCE_DIR}/../../thirdparty/Sophus)
set(Eigen3_DIR ${PROJECT_SOURCE_DIR}/../../thirdparty/g2o/EXTERNAL/eigen-3.2.8)
find_package(catkin REQUIRED)
find_package(cmake_modules REQUIRED)
find_package(Eigen3 REQUIRED)
find_package(Boost REQUIRED COMPONENTS thread)
find_package(OpenCV REQUIRED)
find_package(TBB REQUIRED)
find_package(PCL REQUIRED COMPONENTS common filters)

catkin_package(
  INCLUDE_DIRS include
  LIBRARIES dvo_core dvo_visualization
  DEPENDS OpenCV PCL
)

set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse -msse2 -msse3 -msse4 -ftree-vectorize")

include_directories(
  "include"
  #${Sophus_INCLUDE_DIRS}
  ${Sophus_DIR}
  ${catkin_INCLUDE_DIRS}
  ${EIGEN3_INCLUDE_DIRS}
  ${Boost_INCLUDE_DIRS}
  ${PCL_INCLUDE_DIRS}
  ${OpenCV_INCLUDE_DIRS}
)

add_library(dvo_core 
    src/core/interpolation.cpp
    src/core/intrinsic_matrix.cpp
    src/core/least_squares.cpp
    src/core/math_sse.cpp
    src/core/rgbd_image.cpp
    src/core/rgbd_image_sse.cpp
    src/core/point_selection.cpp
    src/core/surface_pyramid.cpp
    src/core/weight_calculation.cpp
    
src/util/histogram.cpp

src/visualization/camera_trajectory_visualizer.cpp

src/dense_tracking.cpp
src/dense_tracking_impl.cpp
src/dense_tracking_config.cpp
)

target_link_libraries(dvo_core
${PCL_LIBRARIES}
${OpenCV_LIBRARIES}
${catkin_LIBRARIES}
tbb
boost_thread
)

add_library(dvo_visualization
  src/visualization/async_point_cloud_builder.cpp
    src/visualization/point_cloud_aggregator.cpp
#    src/visualization/pcl_camera_trajetory_visualizer.cpp
)

target_link_libraries(dvo_visualization
  dvo_core
)

（2）dvo_ros的CMakeLists.txt
Path：dvo_slam/src/dvo_ros/CMakeLists.txt

cmake_minimum_required(VERSION 2.8.3)
project(dvo_ros)

find_package(catkin REQUIRED COMPONENTS roscpp tf tf_conversions eigen_conversions dynamic_reconfigure dvo_core sensor_msgs image_transport message_filters cv_bridge visualization_msgs interactive_markers pcl_ros)
find_package(cmake_modules REQUIRED)

# dynamic reconfigure
generate_dynamic_reconfigure_options(cfg/CameraDenseTracker.cfg)

catkin_package(
  INCLUDE_DIRS include
  LIBRARIES dvo_ros
)

include_directories(
  "include"
  ${catkin_INCLUDE_DIRS}
)


add_library(${PROJECT_NAME} 
    src/camera_base.cpp
    src/camera_dense_tracking.cpp
        
    src/visualization/ros_camera_trajectory_visualizer.cpp
)

target_link_libraries(${PROJECT_NAME}
  ${catkin_LIBRARIES}
)

add_library(camera_tracker_nodelet
    src/camera_tracker_nodelet.cpp
)

target_link_libraries(camera_tracker_nodelet
    ${PROJECT_NAME}
    ${catkin_LIBRARIES}
)

add_dependencies(camera_tracker_nodelet ${PROJECT_NAME}_gencfg)

add_executable(camera_tracker
     src/camera_tracker_node.cpp
)

target_link_libraries(camera_tracker
    ${PROJECT_NAME}
    ${catkin_LIBRARIES}
)

add_dependencies(camera_tracker ${PROJECT_NAME}_gencfg)

（3）dvo_slam的CMakeLists.txt
Path：dvo_slam/src/dvo_slam/CMakeLists.txt

cmake_minimum_required(VERSION 2.8.3)
project(dvo_slam)

#set(BUILD_SHARED_LIBS OFF)
set(G2O_DIR ${PROJECT_SOURCE_DIR}/../../thirdparty/g2o/build)
find_package(G2O REQUIRED)

LIST(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake_modules")
find_package(catkin REQUIRED COMPONENTS roscpp dynamic_reconfigure dvo_core dvo_ros sensor_msgs image_transport message_filters cv_bridge visualization_msgs interactive_markers)
find_package(cmake_modules REQUIRED)
find_package(SuiteSparse REQUIRED)
find_package(OpenGL)

IF(G2O_FOUND)
message(STATUS "found g2o ${G2O_FOUND}")
ELSE()
message(ERROR "g2o missing")
ENDIF()

# dynamic reconfigure
generate_dynamic_reconfigure_options(cfg/GraphVisualizer.cfg cfg/KeyframeSlam.cfg)

catkin_package(
  INCLUDE_DIRS include
  LIBRARIES dvo_slam
  DEPENDS G2O 
  DEPENDS dvo_core
)

set(Sophus_DIR ${PROJECT_SOURCE_DIR}/../../thirdparty/Sophus)

include_directories(
  "include"
  ${G2O_INCLUDE_DIRS}
  ${dvo_slam_INCLUDE_DIRS}
  ${catkin_INCLUDE_DIRS}
  ${CSPARSE_INCLUDE_DIR}
  ${Sophus_INCLUDE_DIRS}
  ${Sophus_DIR}
)
message("${Sophus_INCLUDE_DIRS} " ${Sophus_INCLUDE_DIRS})

# messages
#rosbuild_genmsg()

# VTK library
#find_package(VTK REQUIRED)
#include_directories(${VTK_INCLUDE_DIRS})

add_library(${PROJECT_NAME} 
src/keyframe_tracker.cpp
  src/keyframe_graph.cpp
  src/keyframe_constraint_search.cpp
  src/camera_keyframe_tracking.cpp

  src/config.cpp

  src/tracking_result_evaluation.cpp
  src/local_map.cpp
  src/local_tracker.cpp

  src/constraints/constraint_proposal.cpp
  src/constraints/constraint_proposal_voter.cpp
  src/constraints/constraint_proposal_validator.cpp

  src/serialization/map_serializer_interface.cpp
  src/serialization/map_serializer.cpp

  src/visualization/graph_visualizer.cpp

  src/types.cpp
)

target_link_libraries(${PROJECT_NAME}
  ${G2O_LIBRARIES}
  ${catkin_LIBRARIES}
  ${OPENGL_LIBRARIES}
)

#add_executable(camera_keyframe_tracker
#  src/camera_keyframe_tracker_node.cpp
#)

#target_link_libraries(camera_keyframe_tracker
#  ${PROJECT_NAME}
#)

add_dependencies(${PROJECT_NAME} ${PROJECT_NAME}_gencfg)

（4）dvo_benchmark的CMakeLists.txt
Path：dvo_slam/src/dvo_benchmark/CMakeLists.txt

cmake_minimum_required(VERSION 2.8.3)

project(dvo_benchmark)

#set(BUILD_SHARED_LIBS OFF)

set(G2O_DIR ${PROJECT_SOURCE_DIR}/../../thirdparty/g2o/build)
find_package(G2O REQUIRED)

LIST(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake_modules")

find_package(catkin REQUIRED COMPONENTS roscpp dvo_core dvo_ros dvo_slam dynamic_reconfigure)
find_package(cmake_modules REQUIRED)

catkin_package(
  INCLUDE_DIRS include
  LIBRARIES dvo_benchmark
)

include_directories(
  "include"
  ${catkin_INCLUDE_DIRS}
  ${G2O_INCLUDE_DIRS}
)


add_executable(benchmark
     src/benchmark_slam.cpp
)

#add_executable(experiment
#     src/experiment.cpp
#)

target_link_libraries(benchmark
#  vtkCommon
  ${G2O_LIBRARIES}
  ${catkin_LIBRARIES}
)

# create output directory - just convenience for our users
#file(MAKE_DIRECTORY "${PROJECT_SOURCE_DIR}/output")
#file(MAKE_DIRECTORY "${PROJECT_SOURCE_DIR}/output/video")

install(DIRECTORY launch/
DESTINATION ${CATKIN_PACKAGE_SHARE_DESTINATION}/launch

5. 编译dvo_slam

按以下顺序编译dvo_slam

catkin_make -DCATKIN_WHITELIST_PACKAGES=dvo_core -DCMAKE_BUILD_TYPE=Release
catkin_make -DCATKIN_WHITELIST_PACKAGES=dvo_ros -DCMAKE_BUILD_TYPE=Release
catkin_make -DCATKIN_WHITELIST_PACKAGES=dvo_slam -DCMAKE_BUILD_TYPE=Release
catkin_make -DCATKIN_WHITELIST_PACKAGES=dvo_benchmark -DCMAKE_BUILD_TYPE=Release

遇到问题：

/home/xxx/workspace/openslam_nav/src/dvo_slam/src/dvo_ros/src/camera_tracker_nodelet.cpp:24:24: error: expected constructor, destructor, or type conversion before ‘(’ token(’
PLUGINLIB_DECLARE_CLASS(dvo_ros, camera_tracker, dvo_ros::CameraTrackerNodelet, nodelet::Nodelet);

解决办法：
dvo_slam/src/dvo_ros/src/camera_tracker_nodelet.cpp +24

PLUGINLIB_EXPORT_CLASS(dvo_ros::CameraTrackerNodelet, nodelet::Nodelet)

6. 运行dvo_slam

roslaunch launch/benchmark.launch dataset:=<RGBD dataset folder>
eg:
roslaunch launch/benchmark.launch dataset:=/home/xxx/Download/bags/RGBD-TUM/tgz/rgbd_dataset_freiburg1_360
rviz -f world

在这里插入图片描述

Nie_Xun
关注
  • 1
    点赞
  • 踩
  • 9
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
医学影像增强：空间域方法与频域方法等
LiRongLu_的博客
04-22 159
这篇博客全面综述了医学影像图像增强技术，尤其是空间域和频域方法。文章首先解释了医学影像的重要性以及图像增强在提高诊断准确性中的作用。接着，深入探讨了空间域方法，包括直方图均衡化、对比度限制自适应直方图均衡化(CLAHE)等技术，以及它们如何改善图像质量，突出重要细节。此外，也分析了频域方法，例如傅里叶变换和小波变换等，在去噪和边缘增强方面的应用。文章评估了这些技术在不同情况下的效果，并讨论了它们在实际临床应用中的潜力和挑战。最后，提出了未来研究方向，强调了结合多种技术和人工智能算法的重要性。
ORB-SLAM3 安装（详细）
07-12
ORB-SLAM3 安装（详细）
SLAM 十四讲 课件 slam
04-23
SLAM 十四讲 课件 slam
dre_slam安装过程中用到的yolo3配置文件
08-17
dre_slam安装过程中用到的yolo3配置文件 对应步骤为： wget https://pjreddie.com/media/files/yolov3.weights wget https://github.com/pjreddie/darknet/blob/master/cfg/yolov3.cfg?raw=true -O ./yolov3.cfg wget https://github.com/pjreddie/darknet/blob/master/data/coco.names?raw=true -O ./coco.names
Matlab实现SLAM算法
06-14
A solution to the simultaneous localization and map building (SLAM) problem 附matlab代码
DVO-SLAM详解
snnily的博客
04-15 8323
一、整体框架本系列的文章主要解释DVO-SLAM的Tracking和闭环检测部分,首先从benchmark_slam.cpp这个文件入手，整个算法框架如下：配置好相关参数后，程序循环读取数据集中的图片，开始运行算法，如下图所示：整个算法流程如下：（1） 配置相关参数，读取所有的rgb图及其对应深度图的保存路径（2） 创建KeyframeTracker对象（3） 循环读取数据    （3.1) 加载...
Cmake3.5.1和Eigen3.2.8最新版
04-15
Cmake3.5.1和Eigen3.2.8最新版（2016）可用于SDM算法
DVO-slam原理介绍
weixin_41469272的博客
11-05 1461
文章目录1、误差约束公式2、权重最小二乘迭代求解(IRLS)：2.1 论文1中的迭代公式：论文1中只说明了光度误差：2.2 论文2中的迭代公式3、关键帧的选择 参考论文1：Robust Odometry Estimation for RGB-D Cameras 参考论文2：Dense Visual SLAM for RGB-D Cameras 1、误差约束公式 基于直接法的稠密建图，误差采用两部分： 灰度误差和深度误差： 灰度误差： 灰度误差表示匹配点在图像中被观测到的光度误差，误差方程定义如下： 深度误
RGBDSLAM安装与使用
chenxingwangzi的博客
10-14 3475
对于RGBDSLAM我看已经有很多文章了，而且还有大神逐步分析了RGBDSLAM的步骤和原理，对此我就不多赘述了，一会上链接大家共同学习一下。前期准备 源码戳这里 ，作者十分负责的写了很长的readme简直业界良心，在里面的Installation中很详细的写了代码的具体安装步骤。 这里我要补充两点： （1）因为我并没有配过GPU所以在源码的cmakelist中的 （ps:博主目录~/rgbd
SLAM14讲 ch6 g2o曲线拟合程序问题-G2OConfig.cmake
Aflws的博客
12-16 1795
SLAM14讲 ch6 g2o曲线拟合程序问题-G2OConfig.cmake Could not find a package configuration file provided by "G2O" with any of the following names: G2OConfig.cmake g2o-config.cmake 这一讲包含了一个用g2o库进行曲线拟合的实例，但是在按照书中实际步骤实际运行发现了几个问题。 亦或者出现这么一个尴尬的场面：利用高博的代码编译可以...
Ubuntu 16.04 下 RGBDSLAM-v2 的安装和使用
CG-SpatialAI
10-14 3541
文章目录简介运行环境安装依赖编译 rgbdslam_v2运行RGB-D 数据集RGB-D 相机参考 简介 RGBD-SLAM-v2 is a state-of-the-art SLAM system for RGB-D cameras, e.g., the Microsoft Kinect. You can use it to create highly accurate 3D point clo...
为水下机器人的单目SLAM实现与评估
08-24
为水下机器人的单目SLAM实现与评估 - 多种slam实现方式评估
RGBD_slam.rar_RGBD_RGBD-slam 特征检测_rgbd slam_slam
09-14
简单的RGBD-slam程序，包含特征点检测，特征描述，点云地图的生成等。
用python学习rgbd-slam系列
01-02
用python学习rgbd-slam系列
SLAM-基于线特征的RGBD-SLAM算法实现-支持与点特征融合-优质项目实战.zip
02-23
SLAM_基于线特征的RGBD-SLAM算法实现_支持与点特征融合_优质项目实战
用python学习rgbd-slam系列.zip
02-22
用python学习rgbd-slam系列
ORB-SLAM2-RGBD-DENSE-MAP-data.tar
08-10
ORB-SLAM2-RGBD-DENSE-MAP-data，D435i相机发出的rgb图和对其的深度图，以及对应的相机内参，数据格式是rosbag。
【深度学习】烟雾和火焰数据集，野外数据集，超大量数据集，目标检测，YOLOv5
最新发布
q742971636的博客
04-23 355
标注了2w张数据集，是目标检测yolo格式的，有火焰、烟雾两个目标。
三个目前主流的计算机视觉软件
地球空间
04-19 800
这些软件在计算机视觉领域都有着重要的地位，它们提供了一系列的工具和库，使得研究人员和开发者能够构建复杂的视觉系统。尽管如此，将计算机视觉的能力直接与人类的视觉能力进行比较并不完全准确，因为它们的工作原理和处理信息的方式存在根本差异。1. **图像识别**：计算机可以识别图像中的物体、场景和活动，准确率在某些情况下甚至超过了人类。3. **语义分割**：将图像中的每个像素分配到特定的类别，用于理解图像的详细内容。5. **图像生成**：生成逼真的图像，如通过GANs（生成对抗网络）。
orb-slam3运行rgbd+imu
08-25
总之，要在ORB-SLAM3中运行RGBD IMU模式，您需要正确安装ORB-SLAM3、配置相关运行环境，并根据官方文档或示例代码了解如何使用RGBD-inertial模式和对应的ROS接口。<span class="em">1</span><span class="em">2...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值