ewenwan / orb_slam2_ssd_semantic Goto Github PK

动态语义SLAM 目标检测+VSLAM+光流/多视角几何动态物体检测+octomap地图+目标数据库

C++ 95.17% CMake 0.63% Python 1.40% C 2.80%

semantic-slam dynamic-slam octomap object-detection

orb_slam2_ssd_semantic's Issues

编译过程中的一些问题

你好，你的工作看起来非常棒！但是我在编译过程中遇到了一些问题，好像 VTK，ORBVocabulary和PCL好像都出现了问题，能帮我看一下吗？谢谢你哟～

from /home/jin/ORB_SLAM2_SSD_Semantic/perfect/src/pointcloudmapping.cc:8:
/usr/include/vtk-6.2/vtkAtomicInt.h:307:28: error: reference to ‘detail’ is ambiguous
 class vtkAtomicInt: public detail::vtkAtomicIntImpl<T>

from /home/jin/ORB_SLAM2_SSD_Semantic/perfect/src/pointcloudmapping.cc:8:
/usr/include/vtk-6.2/vtkAtomicInt.h:307:36: error: expected ‘{’ before ‘vtkAtomicIntImpl’
 class vtkAtomicInt: public detail::vtkAtomicIntImpl<T>
                                    ^
/usr/include/vtk-6.2/vtkAtomicInt.h:307:52: error: expected initializer before ‘<’ token
 class vtkAtomicInt: public detail::vtkAtomicIntImpl<T>

from /home/jin/ORB_SLAM2_SSD_Semantic/perfect/src/pointcloudmapping.cc:8:
/usr/include/vtk-6.2/vtkObjectBase.h:165:30: error: field ‘ReferenceCount’ has incomplete type ‘vtkAtomicInt<int>’
   vtkAtomicInt<vtkTypeInt32> ReferenceCount;

/home/jin/ORB_SLAM2_SSD_Semantic/perfect/src/System.cc: In constructor ‘ORB_SLAM2::System::System(const string&, const string&, ORB_SLAM2::System::eSensor, bool)’:
/home/jin/ORB_SLAM2_SSD_Semantic/perfect/src/System.cc:129:32: error: ‘ORB_SLAM2::ORBVocabulary {aka class DBoW2::TemplatedVocabulary<cv::Mat, DBoW2::FORB>}’ has no member named ‘loadFromBinaryFile’
       bVocLoad = mpVocabulary->loadFromBinaryFile(strVocFile);//bin格式打开

from /home/jin/ORB_SLAM2_SSD_Semantic/perfect/src/MergeSG.cc:14:
/usr/local/include/pcl-1.8/pcl/segmentation/euclidean_cluster_comparator.h:266:13: error: cannot bind ‘uint32_t {aka unsigned int}’ lvalue to ‘std::set<unsigned int>::value_type&& {aka unsigned int&&}’
             exclude_labels_->insert (i);

关于semantic的数据关联

很棒的工作！有semantic mapping的内容但是semantic localization好像还没有融入
看到作者给出了数据关联的思路

使用自己的数据集和目标检测模型运行代码

大家好，请问各位有没有尝试过使用自己的数据集训练出的目标检测模型来运行这套代码呢，大致的修改思路能否讨论一下，如果有做类似工作的也欢迎微信交流（vx: MonsyC），感谢！

运行1.0版本pcl显示错误

您好，我再尝试复现你的1.0版本时，一开始可以显示pcl点云，但是只显示第一次识别后就无法继续运行。我是pcl 1.8.1版本。

我调试问题是在这里只能循环两次，然后线程就死了好像。请问您知道为什么吗。或者请问如何在2.0版本中显示pcl点云信息，谢谢

hidden symbol

/usr/bin/ld: ../Examples/RGB-D/rgbd_tum: hidden symbol `__cpu_indicator_init' in /usr/lib/gcc/x86_64-linux-gnu/5/libgcc.a(cpuinfo.o) is referenced by DSO
/usr/bin/ld: final link failed: Bad value

cmakelist里面把和图漾相机的注释了，最后一步出现这个问题

编译已经快完成了，但是出现：
CMakeFiles/rgbd_ty.dir/Examples/RGB-D/my_rgbd.cc.o：在函数‘frameHandler(TY_FRAME_DATA*, void*)’中：
my_rgbd.cc:(.text+0x1086)：对‘TYRegisterWorldToColor2’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘ncnn::Extractor::extract(char const*, ncnn::Mat&)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘ncnn::Net::register_custom_layer(char const*, ncnn::Layer* (*)())’未定义的引用
collect2: error: ld returned 1 exit status
CMakeFiles/rgbd_ty.dir/build.make:361: recipe for target '../Examples/RGB-D/rgbd_ty' failed
make[2]: *** [../Examples/RGB-D/rgbd_ty] Error 1
CMakeFiles/Makefile2:104: recipe for target 'CMakeFiles/rgbd_ty.dir/all' failed
make[1]: *** [CMakeFiles/rgbd_ty.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
这样的问题，请问怎么解决
我还尝试把不使用图漾相机的编译，但是后面两个对ncnn的未定义引用还是存在，能请大神帮忙看下吗

运行过程中追踪部分出现段错误

`ORB特征提取参数 ORB Extractor Parameters:
-- 每幅图像特征点数量 Number of Features: 1000
-- 金字塔层数Scale Levels: 8
-- 金字塔尺度Scale Factor: 1.2
-- 初始快速角点法阈值 Initial Fast Threshold: 20
-- 最小阈值 Minimum Fast Threshold: 7

深度图阈值 Depth Threshold (Close/Far Points): 2.98842

Start processing sequence ...
Images in the sequence: 827

Light Tracking homo not working because Tracking is not initialized...
新地图创建成功 new map ,具有地图点数 : 718 地图点 points
Tracking homo faild...
receive a keyframe, id = 1
段错误 (核心已转储)
`
can anyone please help me with this problem about tracking? Thanks!!

make编译出现的问题“CMakeFiles/rgbd_tum.dir/build.make:404: recipe for target '../Examples/RGB-D/rgbd_tum' failed”求大神帮忙看一眼问题出在哪里

../lib/libORB_SLAM2_pc.so：对‘vkFreeCommandBuffers’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkBindBufferMemory’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdCopyBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetImageMemoryRequirements’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyInstance’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateSampler’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::setSourceEntryPoint(char const*)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkFreeMemory’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘ncnn::Extractor::extract(char const*, ncnn::Mat&)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateDescriptorSetLayout’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyCommandPool’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkEnumerateInstanceExtensionProperties’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkWaitForFences’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetPhysicalDeviceMemoryProperties’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetInstanceProcAddr’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroySemaphore’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateImageView’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdBindDescriptorSets’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkBeginCommandBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyDevice’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreatePipelineLayout’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateComputePipelines’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::getInfoLog()’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkAllocateMemory’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateInstance’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkResetCommandBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyPipelineLayout’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::parse(TBuiltInResource const*, int, EProfile, bool, bool, EShMessages, glslang::TShader::Includer&)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyFence’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyDescriptorPool’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdCopyImage’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetBufferMemoryRequirements’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkAllocateDescriptorSets’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyPipeline’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyImageView’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkEnumeratePhysicalDevices’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::addProcesses(std::vector<std::__cxx11::basic_string<char, std::char_traits, std::allocator >, std::allocator<std::__cxx11::basic_string<char, std::char_traits, std::allocator > > > const&)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdCopyImageToBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdPipelineBarrier’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetPhysicalDeviceQueueFamilyProperties’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::TShader(EShLanguage)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkUnmapMemory’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetDeviceQueue’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateImage’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdCopyBufferToImage’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetDeviceProcAddr’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkMapMemory’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkResetFences’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::GlslangToSpv(glslang::TIntermediate const&, std::vector<unsigned int, std::allocator >&, glslang::SpvOptions*)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroySampler’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateDevice’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateDescriptorPool’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyShaderModule’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::getInfoDebugLog()’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkEnumerateDeviceExtensionProperties’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyDescriptorSetLayout’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::setStringsWithLengths(char const* const*, int const*, int)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdDispatch’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkBindImageMemory’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkQueueSubmit’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkInvalidateMappedMemoryRanges’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::setEntryPoint(char const*)’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateFence’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateShaderModule’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkGetPhysicalDeviceProperties’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdPushConstants’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::TShader::~TShader()’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkFlushMappedMemoryRanges’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkAllocateCommandBuffers’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkDestroyImage’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::FinalizeProcess()’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkEndCommandBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkFreeDescriptorSets’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateCommandPool’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘glslang::InitializeProcess()’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkUpdateDescriptorSets’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateBuffer’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCreateSemaphore’未定义的引用
../lib/libORB_SLAM2_pc.so：对‘vkCmdBindPipeline’未定义的引用
collect2: error: ld returned 1 exit status
CMakeFiles/rgbd_tum.dir/build.make:404: recipe for target '../Examples/RGB-D/rgbd_tum' failed
make[2]: *** [../Examples/RGB-D/rgbd_tum] Error 1
CMakeFiles/Makefile2:104: recipe for target 'CMakeFiles/rgbd_tum.dir/all' failed
make[1]: *** [CMakeFiles/rgbd_tum.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2

运行时无法显示八叉树地图

您好！您的这个工程十分优秀！从中我学习到了很多知识。但我在运行这个程序时遇到了一些问题，希望能得到您的帮助。在程序开始启动时，终端显示为第一张图片；当点击按钮切换到八叉树地图时，图像和地图窗口都卡住，不能切换到八叉树地图，终端报错如第二张图片所示。想问一下如何解决这个问题呢？十分感谢！

some mistakes

Hello, when i compile the project, there is an error of " *** No rule to make target '/usr/lib/x86_64-linux-gnu/mesa-egl/libEGL.so', needed by '../lib/libORB_SLAM2_pc.so'。STOP ", and what should i do? Thanks very much

小觅标准版相机可以用这套代码吗？

运行中遇到一些问题

你好，我在复现了你的代码之后，尝试了自己的版本，但是在利用ROS运行时，整个轨迹相对于grid（）绘制的网格，是斜向下的，没有和地面平行，请问你遇到过吗？ @Ewenwan

/mobilenetv2_ssdlite_voc.param failed

我的显示如下：

加载 ORB词典. This could take a while...
empty line at 1082073
数据库载入时间 Vocabulary loaded in 6.48s
fopen /home/ewenwan/ewenwan/project/ORB_SLAM2-src-flow--geom-sem-octo/Thirdparty/ncnn/model/mobilenetv2_ssdlite_voc.param failed
fopen /home/ewenwan/ewenwan/project/ORB_SLAM2-src-flow--geom-sem-octo/Thirdparty/ncnn/model/mobilenetv2_ssdlite_voc.bin failed

ncnn已经安装了啊，请问怎么解决？

编译过程中的错误

[ 70%] Building CXX object CMakeFiles/ORB_SLAM2_pc.dir/src/Geometry.cc.o
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc: In member function ‘void DynaSLAM::Geometry::GeometricModelUpdateDB(const ORB_SLAM2::Frame&)’:
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:75:22: error: ‘const class ORB_SLAM2::Frame’ has no member named ‘mIsKeyFrame’
if (currentFrame.mIsKeyFrame)// 是关键帧才更新
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc: In member function ‘std::vectorDynaSLAM::Geometry::DynKeyPoint DynaSLAM::Geometry::ExtractDynPoints(const std::vector<ORB_SLAM2::Frame>&, const ORB_SLAM2::Frame&)’:
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:170:38: error: ‘class ORB_SLAM2::Frame’ has no member named ‘mImDepth’
const float d = refFrame.mImDepth.at(v,u);// 深度值
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:170:50: error: expected primary-expression before ‘float’
const float d = refFrame.mImDepth.at(v,u);// 深度值
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:168:26: warning: unused variable ‘v’ [-Wunused-variable]
const float &v = kp.pt.y;
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:169:26: warning: unused variable ‘u’ [-Wunused-variable]
const float &u = kp.pt.x;
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:354:46: error: ‘const class ORB_SLAM2::Frame’ has no member named ‘mImDepth’
const float d = currentFrame.mImDepth.at(y,x);// 当前�
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:354:58: error: expected primary-expression before ‘float’
const float d = currentFrame.mImDepth.at(y,x);// 当前�
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:404:41: error: ‘const class ORB_SLAM2::Frame’ has no member named ‘mImDepth’
float _d = currentFrame.mImDepth.at(y,x);// 对应的深
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:404:53: error: expected primary-expression before ‘float’
float _d = currentFrame.mImDepth.at(y,x);// 对应的深
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:402:21: warning: unused variable ‘x’ [-Wunused-variable]
int x = (int)matCurrentFrame.at(i,0) + (int)u1.at
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:403:21: warning: unused variable ‘y’ [-Wunused-variable]
int y = (int)matCurrentFrame.at(i,1) + (int)u1.at
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:420:21: warning: unused variable ‘xIndex’ [-Wunused-variable]
int xIndex = minIdx.x;// 在关键点前方，靠的较近，��
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:448:46: error: ‘const class ORB_SLAM2::Frame’ has no member named ‘mImDepth’
cv::Mat patch = currentFrame.mImDepth.rowRange(yIni,yEnd).colRa
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc: In member function ‘bool DynaSLAM::Geometry::IsInFrame(const float&, const float&, const ORB_SLAM2::Frame&)’:
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:592:43: error: ‘const class ORB_SLAM2::Frame’ has no member named ‘mImDepth’
return (x > (mDmax + 1) && x < (Frame.mImDepth.cols - mDmax - 1) && y > (mD
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:592:103: error: ‘const class ORB_SLAM2::Frame’ has no member named ‘mImDepth’
(Frame.mImDepth.cols - mDmax - 1) && y > (mDmax + 1) && y < (Frame.mImDepth.ro
^
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc: In member function ‘cv::Mat DynaSLAM::Geometry::RegionGrowing(const cv::Mat&, int&, int&, const float&)’:
/home/y/ORB_SLAM2_SSD_Semantic-master/ORB_SLAM2_SSD_Semantic-master/perfect/src/Geometry.cc:636:45: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
while(pixdist < reg_maxdist && reg_size < im.total())
^
CMakeFiles/ORB_SLAM2_pc.dir/build.make:542: recipe for target 'CMakeFiles/ORB_SLAM2_pc.dir/src/Geometry.cc.o' failed
make[2]: *** [CMakeFiles/ORB_SLAM2_pc.dir/src/Geometry.cc.o] Error 1
CMakeFiles/Makefile2:104: recipe for target 'CMakeFiles/ORB_SLAM2_pc.dir/all' failed
make[1]: *** [CMakeFiles/ORB_SLAM2_pc.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2

论文链接

你好，论文的链接能否发一下？

/usr/bin/ld: 找不到....

/usr/bin/ld: 找不到 -lvtkxdmf2
/usr/bin/ld: 找不到 -lvtkGUISupportQt
/usr/bin/ld: 找不到 -lvtkCommonExecutionModel
/usr/bin/ld: 找不到 -lvtkCommonDataModel
/usr/bin/ld: 找不到 -lvtkCommonMath
/usr/bin/ld: 找不到 -lvtkCommonCore
/usr/bin/ld: 找不到 -lvtkCommonMisc
/usr/bin/ld: 找不到 -lvtkCommonSystem
/usr/bin/ld: 找不到 -lvtkCommonTransforms
/usr/bin/ld: 找不到 -lvtkFiltersExtraction
/usr/bin/ld: 找不到 -lvtkFiltersCore
/usr/bin/ld: 找不到 -lvtkFiltersGeneral
/usr/bin/ld: 找不到 -lvtkCommonComputationalGeometry
/usr/bin/ld: 找不到 -lvtkFiltersStatistics
/usr/bin/ld: 找不到 -lvtkImagingFourier
/usr/bin/ld: 找不到 -lvtkImagingCore
/usr/bin/ld: 找不到 -lvtkInteractionStyle
/usr/bin/ld: 找不到 -lvtkFiltersSources
/usr/bin/ld: 找不到 -lvtkRenderingCore
/usr/bin/ld: 找不到 -lvtkCommonColor
/usr/bin/ld: 找不到 -lvtkFiltersGeometry
/usr/bin/ld: 找不到 -lvtkRenderingOpenGL
/usr/bin/ld: 找不到 -lvtkImagingHybrid
/usr/bin/ld: 找不到 -lvtkIOImage
/usr/bin/ld: 找不到 -lvtkIOCore
/usr/bin/ld: 找不到 -lvtkFiltersParallelGeometry
/usr/bin/ld: 找不到 -lvtkParallelMPI
/usr/bin/ld: 找不到 -lvtkParallelCore
/usr/bin/ld: 找不到 -lvtkIOLegacy
/usr/bin/ld: 找不到 -lvtkImagingColor
/usr/bin/ld: 找不到 -lvtkIOGeoJSON
/usr/bin/ld: 找不到 -lvtkFiltersParallelImaging
/usr/bin/ld: 找不到 -lvtkFiltersImaging
/usr/bin/ld: 找不到 -lvtkImagingGeneral
/usr/bin/ld: 找不到 -lvtkImagingSources
/usr/bin/ld: 找不到 -lvtkFiltersParallel
/usr/bin/ld: 找不到 -lvtkFiltersModeling
/usr/bin/ld: 找不到 -lvtkRenderingFreeType
/usr/bin/ld: 找不到 -lvtkIOParallelNetCDF
/usr/bin/ld: 找不到 -lvtkWrappingTools
/usr/bin/ld: 找不到 -lvtkIOInfovis
/usr/bin/ld: 找不到 -lvtkIOXML
/usr/bin/ld: 找不到 -lvtkIOGeometry
/usr/bin/ld: 找不到 -lvtkIOXMLParser
/usr/bin/ld: 找不到 -lvtkInfovisCore
/usr/bin/ld: 找不到 -lvtkIOXdmf2
/usr/bin/ld: 找不到 -lvtkIOParallel
/usr/bin/ld: 找不到 -lvtkIONetCDF
/usr/bin/ld: 找不到 -lvtkFiltersParallelMPI
/usr/bin/ld: 找不到 -lvtkFiltersGeneric
/usr/bin/ld: 找不到 -lvtkInfovisBoostGraphAlgorithms
/usr/bin/ld: 找不到 -lverdict
/usr/bin/ld: 找不到 -lvtkIOLSDyna
/usr/bin/ld: 找不到 -lvtkRenderingLOD
/usr/bin/ld: 找不到 -lvtkFiltersTextu

光流法去除动态点的时候运行结果没有给的图片结果那么好

perfect文件怎么编译？

我按照您给的编译方法，在make这一步出现了这个问题：
CMakeFiles/ORB_SLAM2_pc.dir/build.make:518: recipe for target 'CMakeFiles/ORB_SLAM2_pc.dir/src/pointcloudmapping.cc.o' failed
make[2]: *** [CMakeFiles/ORB_SLAM2_pc.dir/src/pointcloudmapping.cc.o] Error 1
CMakeFiles/Makefile2:104: recipe for target 'CMakeFiles/ORB_SLAM2_pc.dir/all' failed
make[1]: *** [CMakeFiles/ORB_SLAM2_pc.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
后来我去看，发现CMakeFiles/ORB_SLAM2_pc.dir/src文件夹中并没有pointcloudmapping.cc.o这个文件，网上的方法都试过了，但是还是没能解决。我在我的电脑上已经编译通过ORB SLAM2。

我也是同样出现这个错误

Originally posted by @lsj1005 in #10 (comment)

请问你的opencv pcl vtk 配置是怎么样的？我是3.4 1.9 6.2

在perfect中编译时，flann报错了一大段，麻烦问下怎么可以解决呢

我在编译的时候，flann报错了很多，/usr/include/flann/algorithms/all_indices.h:129:98: error: expected primary-expression before ‘const’
ate_index_(flann::Matrix data, const flann::IndexParams& params, const Distance& distance,
类似于这种错误

安装了ncnn，库的目录也导入了，为什么会出现这个问题呢？

error: ‘ncnn’ does not name a type; did you mean ‘nanl’?
ncnn::Net * det_net_ptr;// 检测网络指针

这是跑数据集的结果，请问如何得到带有标签的识别结果图？还有如何用修改用kinectV2相机实时运行呢？谢谢大家了

请问如何保存octomap.ot 文件？

我想尝试编译运行，有详细的文档教程吗

我已经有ORBslam下的依赖环境了

How to compile

求写个在某一版本ubuntu完整编译过程的文档

运行过程中的一些问题

在我按照大家的一些建议下终于编译好了这个文件，但是在运行的时候却出现了以下的问题，非常希望您可以给出解答。
第一个错误：

在我运行的时候缺少了libopencv_core_3.so.3.2 ,我在我的OPENCV安装包里面也没有找到这个文件。我看网上说适合ROS相关的？那么应该应该如何改正

第二个错误：

程序执行到System.cc 第153行的时候：
mpRunDetect = make_shared();// 目标检测智能指针对象====
出现了 malloc(): memory corruption 错误。也不知道怎么改正，谢谢！

my_rgbd.cc:(.text+0x4ca5)：对‘TYRegisterWorldToColor2’未定义的引用

CMakeFiles/rgbd_ty.dir/Examples/RGB-D/my_rgbd.cc.o：在函数‘frameHandler(TY_FRAME_DATA*, void*)’中：
my_rgbd.cc:(.text+0x4ca5)：对‘TYRegisterWorldToColor2’未定义的引用
collect2: error: ld returned 1 exit status
CMakeFiles/rgbd_ty.dir/build.make:391: recipe for target '../Examples/RGB-D/rgbd_ty' failed
make[2]: *** [../Examples/RGB-D/rgbd_ty] Error 1
CMakeFiles/Makefile2:104: recipe for target 'CMakeFiles/rgbd_ty.dir/all' failed
make[1]: *** [CMakeFiles/rgbd_ty.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2

bug

The file in the picture is not stored in the repository. Where should I download it？

支持导航和避障吗

你好，请问一下，这个支持支持导航和避障吗？小白求教

perfect运行Monocular段错误

lenovo@R9000-3:/opt/c_project/ORB_SLAM2_SSD_Semantic$ ./Examples/Monocular/mono_euroc /opt/oRB_SLAM2/Vocabulary/ORBvoc.bin ./Examples/Monocular/EuRoC.yaml /opt/slam_video/V1_01_easy/mav0/cam0/data/ ./Examples/Monocular/EuRoC_TimeStamps/V101.txt

ORB-SLAM2 单目双目深度相机 SLAM
under certain conditions. See LICENSE.txt.

输入相机为: Monocular

加载 ORB词典. This could take a while...
数据库载入时间 Vocabulary loaded in 0.23s

相机参数 Camera Parameters:
-- fx: 458.654
-- fy: 457.296
-- cx: 367.215
-- cy: 248.375
-- k1: -0.283408
-- k2: 0.0739591
-- p1: 0.00019359
-- p2: 1.76187e-05
-- fps: 20
-- 彩色图通道顺序color order: RGB (ignored if grayscale)

ORB特征提取参数 ORB Extractor Parameters:
-- 每幅图像特征点数量 Number of Features: 1000
-- 金字塔层数Scale Levels: 8
-- 金字塔尺度Scale Factor: 1.2
-- 初始快速角点法阈值 Initial Fast Threshold: 20
-- 最小阈值 Minimum Fast Threshold: 7

Start processing sequence ...
Images in the sequence: 2912

新地图 New Map created with 85 points
初始化错误重置 Wrong initialization, reseting...
系统重置 System Reseting
重置局部建图 Reseting Local Mapper... done
重置回环检测 Reseting Loop Closing... done
重置数据库 Reseting Database... done
新地图 New Map created with 111 points
receive a keyframe, id = 2
段错误 (核心已转储)

ewenwan / orb_slam2_ssd_semantic Goto Github PK

orb_slam2_ssd_semantic's Issues

Recommend Projects

Recommend Topics

Recommend Org