zdzhaoyong / gslam Goto Github PK

Program does not run when I enter "gslam Dataset=<dataset_path>/DroneMap/phantom3-village/phantom3-village-kfs/.npudronemap".
In this dataset, how did you generate the 'Plane' in config.cfg? What does plane mean?

Going to add support of DBow, which is header only and no dependency to OpenCV.
The implementation should be more fast and cost smaller memory.

目前采用的ubuntu16.04,编译通过gslam和gslam_dso和gslam_orb。运行
/gslam Dataset=/home/davidhan/GSLAM/doc/Datasets/mono.tummono SLAM=/home/davidhan/GSLAM-ORBSLAM/build/libgslam_orbslam.so ORBVocabularyFile=/home/davidhan/GSLAM-ORBSLAM/data/ORB_New.voc ORBextractor.nFeatures=400的时候出现卡顿，无法线界面上显示pcl的点云。可视化界面右边一直是黑色

When running the test procedure:
$ gslam Act=Tests --gtest_filter=*
The gslam program gives the following output:
[==========] Running 27 tests from 8 test cases.
[----------] Global test environment set-up.
[----------] 1 test from SLAM
[ RUN ] SLAM.LoadRelease
[ OK ] SLAM.LoadRelease (0 ms)
[----------] 1 test from SLAM (0 ms total)

[----------] 6 tests from Svar
[ RUN ] Svar.Int
[ OK ] Svar.Int (0 ms)
[ RUN ] Svar.Double
[ OK ] Svar.Double (0 ms)
[ RUN ] Svar.String
[ OK ] Svar.String (0 ms)
[ RUN ] Svar.Pointer
[ OK ] Svar.Pointer (0 ms)
[ RUN ] Svar.VecParament
[ OK ] Svar.VecParament (0 ms)
[ RUN ] Svar.Parse
Auther of GSLAM is ZhaoYong
Name is ChangedName
FunctionCalled=1
[ OK ] Svar.Parse (0 ms)
[----------] 6 tests from Svar (0 ms total)

[----------] 1 test from Scommand
[ RUN ] Scommand.Regist
[ OK ] Scommand.Regist (0 ms)
[----------] 1 test from Scommand (0 ms total)

[----------] 1 test from HashMapTest
[ RUN ] HashMapTest.MultiThreadReadWrite
[ OK ] HashMapTest.MultiThreadReadWrite (2 ms)
[----------] 1 test from HashMapTest (2 ms total)

[----------] 6 tests from Estimator
[ RUN ] Estimator.HomographyRANSAC
H:,0.998268,1.99992,3.07404,3.99708,5.00007,6.08132,-4.52509e-06,1.78728e-06,1
MidNoise:0.110101,MidError:0.149897
[ OK ] Estimator.HomographyRANSAC (74 ms)
[ RUN ] Estimator.FundamentalSevenPoint
/home/shihl/workspace/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:88: Failure
Expected: (fabs(F[0]-4.81441976)) <= (1e-6), actual: 1.00506e-06 vs 1e-06
/home/shihl/workspace/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:89: Failure
Expected: (fabs(F[1]+8.16978909)) <= (1e-6), actual: 1.07973e-06 vs 1e-06
/home/shihl/workspace/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:90: Failure
Expected: (fabs(F[2]-6.73133404)) <= (1e-6), actual: 1.67578e-06 vs 1e-06
/home/shihl/workspace/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:94: Failure
Expected: (fabs(-9.92570126-F[6])) <= (1e-6), actual: 1.47626e-06 vs 1e-06
[ FAILED ] Estimator.FundamentalSevenPoint (0 ms)
[ RUN ] Estimator.FundamentalEightPoint
[ OK ] Estimator.FundamentalEightPoint (1 ms)
[ RUN ] Estimator.EssentialFivePoint
[ OK ] Estimator.EssentialFivePoint (0 ms)
[ RUN ] Estimator.SE3PlaneRansac
[ OK ] Estimator.SE3PlaneRansac (0 ms)
[ RUN ] Estimator.Triangulate
[ OK ] Estimator.Triangulate (0 ms)
[----------] 6 tests from Estimator (75 ms total)

[----------] 3 tests from GImageTest
[ RUN ] GImageTest.GImageReadThreadSafe
[ OK ] GImageTest.GImageReadThreadSafe (1001 ms)
[ RUN ] GImageTest.CheckGImageType
[ OK ] GImageTest.CheckGImageType (0 ms)
[ RUN ] GImageTest.OpenCVNoCopy
[ OK ] GImageTest.OpenCVNoCopy (237 ms)
[----------] 3 tests from GImageTest (1238 ms total)

[----------] 1 test from Thread
[ RUN ] Thread.ThreadPool
[ OK ] Thread.ThreadPool (71 ms)
[----------] 1 test from Thread (71 ms total)

[----------] 8 tests from Optimizer
[ RUN ] Optimizer.OptimizePoseSimulate
gslam: symbol lookup error: ./libgslam_optimizer.so: undefined symbol: cholmod_start

Is that normal? What is wrong? Why this error comes out during runtime?

Hello, bro. When I run the 'gslam Act=Tests --gtest_filter=', I meet an error in my ubuntu 16.04.
the log as follows:
duke@ubuntu:~$ gslam Act=Tests --gtest_filter=
[==========] Running 27 tests from 8 test cases.
[----------] Global test environment set-up.
[----------] 3 tests from GImageTest
[ RUN ] GImageTest.GImageReadThreadSafe
[ OK ] GImageTest.GImageReadThreadSafe (1003 ms)
[ RUN ] GImageTest.CheckGImageType
[ OK ] GImageTest.CheckGImageType (0 ms)
[ RUN ] GImageTest.OpenCVNoCopy
[ OK ] GImageTest.OpenCVNoCopy (301 ms)
[----------] 3 tests from GImageTest (1304 ms total)

[----------] 6 tests from Estimator
[ RUN ] Estimator.HomographyRANSAC
H:,0.998719,1.99912,3.06695,3.99617,4.99894,6.09501,-7.13773e-06,2.28781e-06,1
MidNoise:0.111859,MidError:0.163132
[ OK ] Estimator.HomographyRANSAC (5 ms)
[ RUN ] Estimator.FundamentalSevenPoint
/home/duke/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:88: Failure
Expected: (fabs(F[0]-4.81441976)) <= (1e-6), actual: 1.00506e-06 vs 1e-06
/home/duke/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:89: Failure
Expected: (fabs(F[1]+8.16978909)) <= (1e-6), actual: 1.07973e-06 vs 1e-06
/home/duke/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:90: Failure
Expected: (fabs(F[2]-6.73133404)) <= (1e-6), actual: 1.67578e-06 vs 1e-06
/home/duke/GSLAM/GSLAM/gslam/Tests/EstimatorTest.cpp:94: Failure
Expected: (fabs(-9.92570126-F[6])) <= (1e-6), actual: 1.47626e-06 vs 1e-06
[ FAILED ] Estimator.FundamentalSevenPoint (0 ms)
[ RUN ] Estimator.FundamentalEightPoint
[ OK ] Estimator.FundamentalEightPoint (0 ms)
[ RUN ] Estimator.EssentialFivePoint
[ OK ] Estimator.EssentialFivePoint (0 ms)
[ RUN ] Estimator.SE3PlaneRansac
[ OK ] Estimator.SE3PlaneRansac (1 ms)
[ RUN ] Estimator.Triangulate
[ OK ] Estimator.Triangulate (0 ms)
[----------] 6 tests from Estimator (6 ms total)

[----------] 8 tests from Optimizer
[ RUN ] Optimizer.OptimizePoseSimulate
[ OK ] Optimizer.OptimizePoseSimulate (57 ms)
[ RUN ] Optimizer.OptimizePnPSimulate
[ OK ] Optimizer.OptimizePnPSimulate (4 ms)
[ RUN ] Optimizer.ICPSimulate
[ OK ] Optimizer.ICPSimulate (4 ms)
[ RUN ] Optimizer.SE3GraphSimulate

I don't know what that mean. So help me solve it if u free please.

hi,thanks your work.I try to test ,but I get a error,could you help me?
robot@robot:~/slam_evaluate/GSLAM$ gslam Act=Tests --gtest_filter=* gslam: error while loading shared libraries: libopencv_videoio.so.3.0: cannot open shared object file: No such file or directory
Thanks!

CMake Error: CMake can not determine linker language for target: gslamDB_tumrgbd
CMake Error: Cannot determine link language for target "gslamDB_tumrgbd".
CMake Error: CMake can not determine linker language for target: gslamDB_cvmono
CMake Error: Cannot determine link language for target "gslamDB_cvmono".
CMake Error: CMake can not determine linker language for target: gslamDB_euroc
CMake Error: Cannot determine link language for target "gslamDB_euroc".
CMake Error: CMake can not determine linker language for target: gslamDB_kitti
CMake Error: Cannot determine link language for target "gslamDB_kitti".
CMake Error: CMake can not determine linker language for target: gslamDB_npudronemap
CMake Error: Cannot determine link language for target "gslamDB_npudronemap".
CMake Error: CMake can not determine linker language for target: gslamDB_npurgbd
CMake Error: Cannot determine link language for target "gslamDB_npurgbd".
CMake Error: CMake can not determine linker language for target: gslamDB_tummono
CMake Error: Cannot determine link language for target "gslamDB_tummono".
-- Generating done
-- Build files have been written to: /home/weismarts/Documents/libs/GSLAM/build
Makefile:1256: recipe for target 'cmake_check_build_system' failed

I have managed to get SVO working on some of the datasets, I was just wondering how I now use the evaluation component?

How to Chineseize the qviz display window？ such as setObjectName("Messenger Topics") , setObjectName("Frame Visualizer") and so on.

Not sure what I did wrong. It can compile and link correctly. But when I tested the program with ./gslam tests --gtest_filter=*, it reported this. I use Mac Catalina 10.15.1. Thanks.

Hi, I got a problem when I run dso on tumrgdb dataset, the command and error output are follow. The qviz windows has been open but the middle main 3D part is black and no picture show.
I am utterly confused that whether this problem is caused I haven't install QT dependence correctly or the gslam code have some problem?
Please help me. Thanks!

➜  Downloads gslam qviz dso play -dataset ./rgbd_dataset_freiburg3_structure_texture_far/.tumrgbd -autostart

=============== PRESET Settings: ===============
FAST settings:
- 5x real-time enforcing
- 800 active points
- 4-6 active frames
- 1-4 LM iteration each KF
- 424 x 320 image resolution
==============================================
main.cpp:72 Success opened dataset ./rgbd_dataset_freiburg3_structure_texture_far/.tumrgbd
K=  525     0 319.5
    0   525 239.5
    0     0     1
using pyramid levels 0 to 3. coarsest resolution: 80 x 60!
Undistorter:
    Camera IN : PinHole:[640,480,525,525,319.5,239.5]
    Camera OUT: PinHole:[640,480,525,525,319.5,239.5]

QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
.....
QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
QFSFileEngine::open: No file name specified
Initialization: keep 9.2% (need 800, have 8729)!
INITIALIZE FROM INITIALIZER (797 pts)!
[1]    83391 segmentation fault (core dumped)  gslam qviz dso play -dataset  -autostart

Enviroment:

OS: ubuntu 20.04 
QT: qt5
opencv: opencv3.2.0
GSLAM: master version in 2021.4.1

Hi Zhao Yong, could you specify the dataset file, if I want to test slam system by using gslam. Thank you.

Hello @zdzhaoyong,
Thank for your great contribution. I want to ask you a question.
Can GSLAM run directly on webcam or RGB-D camera?

I notice that Estimator and Optimizer plug in no longer include in the branch, are they available in a different repo?

Hi, title says all.

However that packages(i.e., libqt4-blah) just changed its names as starting with 'libqt', and ends with '4', which should be formed as libqtblah4.

In other words, It would be better to change Qt installation guides from
Qt : sudo apt-get install build-essential g++ libqt4-core libqt4-dev libqt4-gui qt4-doc qt4-designer

to
Qt : sudo apt-get install build-essential g++ libqtcore4 libqtdev4 libqtgui4 qt4-doc qt4-designer

At the 3D Transformations: SO3, SE3, SIM3 page there is a comparison table. Value for SE3::exp for Sophus is 19.8 but should be 249.2 according to the paper.
Most probably just a copy-paste from the previous line

Hi there:

I download the TUM RGBD Dataset with name "rgbd_dataset_freiburg1_360.tgz"

But I cannot find .tumrgbd file on the uncompressed fold.

Dataset=<dataset_path>/rgbd_dataset_freiburg1_360/.tumrgbd

Thanks for your great works.
I read the paper "GSLAM: A General SLAM Framework and Benchmark"
It indicates the python binding is provided for the modern deep learning platforms and developers prefer python for coding.
But I cannot any python bindings and python files.
is the development of this feature in progress?
Thank you

Svar should improve on the following directions:

• Now a thread safe Svar or SvarWithType is nocopyable since mutex are used, the implementation can be wrapped so that it could be copyable and a clone function is suggested.
• The Svar should better support XML,JSON,YAML formats load and save, as three of them all support tree like data structure. The GetChild,GetChildren are suppose to be added.

Great work!
Just found a minor inconsistency in the bibtex with your arXiv paper when I was citing it.
Suggestion:
title={General SLAM Framework and Benchmark},
->
title={{GSLAM:} {A} General {SLAM} Framework and Benchmark},

Hello,
when I try to compile the project I get that error:

/home/xxx/Bureau/developpement/SLAM/GSLAM/GSLAM/plugins/cfg/../../core/GImage.h: In member function ‘GSLAM::GImage::operator cv::Mat() const’:
/home/xxx/Bureau/developpement/SLAM/GSLAM/GSLAM/plugins/cfg/../../core/GImage.h:281:37: error: no matching function for call to ‘cv::UMatData::UMatData()’
             cv::UMatData* u=new cv::UMatData;

I am on ubuntu 16.04 with Opencv 3.2.

Hi, when I tried to play back dataset it always give the error:

gslam qviz play -dataset ~/SLAM/dataset/KITTI/odometry/sequences/00/stereo.kitti
error:
corrupted size vs. prev_size
Aborted (core dumped)

Any help with this?

How do you run gslam on a video using CVMono?

Hi, thanks for the work! I am wondering whether GSLAM supports inertial sensor, such as the IMU, for VI SLAM such as OKVIS?

When I added this test for GImage, after the destruct function called, the data pointer are not equal with nullptr. That's why?

TEST(GImageTest,CorrectRelease)
{
    GSLAM::GImage img(10,10,GSLAM::GImageType<uchar,1>::Type);
    {
        GSLAM::GImage img2=img;
    }
    EXPECT_FALSE(img.empty());
    EXPECT_EQ(1,*img.refCount);
    EXPECT_TRUE(img.data);
    img.~GImage();
    EXPECT_EQ(NULL,img.refCount);
    EXPECT_TRUE(img.data==nullptr);// FIXME: why not equal nullptr
    EXPECT_TRUE(img.empty());
}

Hello and thank you for your work!

I want to use GSLAM via Python3 and fail to understand the install. I am able to build the project via cmake, but I cannot install via pip3.
When using pip3 install GSLAM I get the following error:
ERROR: Could not find a version that satisfies the requirement GSLAM (from versions: none)
ERROR: No matching distribution found for GSLAM
So then I tried via the provided setup.py file but I cannot find any module to import.

What am I doing wrong and why can I not install via pypy?

Thank you for any help!

I am receiving the following error when compiling:

/home/nick/Code/gslam/GSLAM/GSLAM/gslam/IO/DatasetNPURGBD.cpp:59:37: error: conversion from ‘cv::Mat’ to non-scalar type ‘GSLAM::GImage’ requested
         GSLAM::GImage img=cv::imread(rgb_file);
                           ~~~~~~~~~~^~~~~~~~~~
/home/nick/Code/gslam/GSLAM/GSLAM/gslam/IO/DatasetNPURGBD.cpp:60:39: error: conversion from ‘cv::Mat’ to non-scalar type ‘GSLAM::GImage’ requested
         GSLAM::GImage depth=cv::imread(d_file,cv::IMREAD_UNCHANGED);

If it is helpful this is my CMAKE output:

cmake ..
-- Compiling GSLAM Version 2.4.7.
-- Build type: Release
-- Looking for Q_WS_X11
-- Looking for Q_WS_X11 - found
-- Looking for Q_WS_WIN
-- Looking for Q_WS_WIN - not found
-- Looking for Q_WS_QWS
-- Looking for Q_WS_QWS - not found
-- Looking for Q_WS_MAC
-- Looking for Q_WS_MAC - not found
-- Found Qt4: /usr/bin/qmake-qt4 (found version "4.8.7") 
CMake Warning (dev) at /usr/share/cmake-3.13/Modules/FindOpenGL.cmake:270 (message):
  Policy CMP0072 is not set: FindOpenGL prefers GLVND by default when
  available.  Run "cmake --help-policy CMP0072" for policy details.  Use the
  cmake_policy command to set the policy and suppress this warning.

  FindOpenGL found both a legacy GL library:

    OPENGL_gl_LIBRARY: /usr/lib/x86_64-linux-gnu/libGL.so

  and GLVND libraries for OpenGL and GLX:

    OPENGL_opengl_LIBRARY: /usr/lib/x86_64-linux-gnu/libOpenGL.so
    OPENGL_glx_LIBRARY: /usr/lib/x86_64-linux-gnu/libGLX.so

  OpenGL_GL_PREFERENCE has not been set to "GLVND" or "LEGACY", so for
  compatibility with CMake 3.10 and below the legacy GL library will be used.
Call Stack (most recent call first):
  cmake/PICMake.cmake:572 (find_package)
  cmake/PICMake.cmake:453 (pi_collect_packages)
  CMakeLists.txt:26 (pi_add_target)
This warning is for project developers.  Use -Wno-dev to suppress it.

-- Found OpenGL: /usr/lib/x86_64-linux-gnu/libOpenGL.so   
-- Found OpenCV: /usr/local (found version "4.1.0") 
------------------------------------------
--Qt: VERSION 4
  QT_INCLUDES: /usr/include/qt4
  QT_LIBRARIES: Qt4::QtCore;Qt4::QtGui;Qt4::QtNetwork;Qt4::QtOpenGL;Qt4::QtSvg;Qt4::QtSql;Qt4::QtXml
  QT_DEFINITIONS: -DHAS_QT
------------------------------------------
--OpenGL:
  OPENGL_INCLUDES: /usr/include
  OPENGL_LIBRARIES: /usr/lib/x86_64-linux-gnu/libGL.so;/usr/lib/x86_64-linux-gnu/libGLU.so
  OPENGL_DEFINITIONS: -DHAS_OPENGL
------------------------------------------
--OpenCV: VERSION 4.1.0
  OPENCV_INCLUDES: /usr/local/include/opencv4
  OPENCV_LIBRARIES: opencv_calib3d;opencv_core;opencv_dnn;opencv_features2d;opencv_flann;opencv_gapi;opencv_highgui;opencv_imgcodecs;opencv_imgproc;opencv_ml;opencv_objdetect;opencv_photo;opencv_stitching;opencv_video;opencv_videoio
  OPENCV_DEFINITIONS: -DHAS_OPENCV
------------------------------------------
--System:
  SYSTEM_LIBRARIES: pthread;dl
  SYSTEM_DEFINITIONS: -DHAS_SYSTEM
-- Found Eigen3: /usr/include/eigen3 (Required is at least version "2.91.0") 
-- Could NOT find ceres (missing: CERES_LIBRARY CERES_INCLUDE_DIR) 
-- GLOG_LIBRARIES = /usr/lib/x86_64-linux-gnu/libglog.so
-- Could NOT find CHOLMOD (missing: CHOLMOD_INCLUDE_DIR CHOLMOD_LIBRARIES) 
------------------------------------------
--Eigen3: VERSION 3.3.7
  EIGEN3_INCLUDES: /usr/include/eigen3
  EIGEN3_DEFINITIONS: -DHAS_EIGEN3
------------------------------------------
--Ceres:
------------------------------------------
--GLOG:
  GLOG_INCLUDES: /usr/include
  GLOG_LIBRARIES: /usr/lib/x86_64-linux-gnu/libglog.so
  GLOG_DEFINITIONS: -DHAS_GLOG
------------------------------------------
--Cholmod:
gslam_optimizer aborded since can't find dependency Ceres.
-- The following targets will to be build:
-- LIBS():  libgslam_estimator.so; libgslamDB_euroc.so; libgslamDB_kitti.so; libgslamDB_npudronemap.so; libgslamDB_npurgbd.so; libgslamDB_cvmono.so; libgslamDB_rtm.so; libgslamDB_tummono.so; libgslamDB_tumrgbd.so
-- APPS():  gslam
-- C++ flags (Release):  -std=c++11 -O3 -DNDEBUG
-- Configuring done
-- Generating done
-- Build files have been written to: /home/nick/Code/gslam/GSLAM/build

Hi, I tried to run the following command

gslam qviz orbslam play -dataset ~/dataset/tum_rgbd/rgbd_dataset_freiburg1_360/.tumrgbd
Unable to load plugin orbslam
PATH=:.:/usr/lib/:/usr/lib64/:/usr/local/lib/:/usr/local/lib64/:
main.cpp:23 Svar module "orbslam" not found.
[1]    27754 abort (core dumped)  gslam qviz orbslam play -dataset 

It shows unable to load plugin orbslam. Do I need to install orbslam module individually?

When I tried to compile the project, it couldn't finish.
Looks like it needs QGLWidget and can't find it nor does it exist in the CMakeLists.txt file.

It gave me the following error:

In file included from /path/GSLAM/GSLAM/gslam/GUI/QGLViewer/vec.h:32:0,
from /path/GSLAM/GSLAM/gslam/GUI/QGLViewer/quaternion.h:26,
from /path/GSLAM/GSLAM/gslam/GUI/QGLViewer/keyFrameInterpolator.h:29,
from /path/GSLAM/GSLAM/gslam/GUI/QGLViewer/camera.h:26,
from /path/GSLAM/GSLAM/gslam/GUI/QGLViewer/qglviewer.h:26,
from /path/GSLAM/GSLAM/gslam/GUI/MainWindow.h:20,
from /path/GSLAM/GSLAM/gslam/GUI/MainWindow.cpp:14:
/path/GSLAM/GSLAM/gslam/GUI/QGLViewer/config.h:78:11: fatal error: QGLWidget: No such file or directory
# include <QGLWidget>
compilation terminated.
CMakeFiles/gslam.dir/build.make:86: recipe for target 'CMakeFiles/gslam.dir/GSLAM/gslam/GUI/MainWindow.cpp.o' failed
make[2]: *** [CMakeFiles/gslam.dir/GSLAM/gslam/GUI/MainWindow.cpp.o] Error 1
CMakeFiles/Makefile2:141: recipe for target 'CMakeFiles/gslam.dir/all' failed
make[1]: *** [CMakeFiles/gslam.dir/all] Error 2
Makefile:129: recipe for target 'all' failed
make: *** [all] Error 2

Thanks for your wonderful GSLAM !

Is collaborative SLAM supported or planned ?

I looked for in the API without succes ..

Thanks in advance.

error: conversion from ‘cv::Mat’ to non-scalar type ‘GSLAM::GImage’ requested
GSLAM::GImage img0=cv::imread(dirtop+"/cam0/data/"+nextImage);

Hello ✋ I don't think this is necessary an "issue", but I just noticed that the datasets on http://zhaoyong.adv-ci.com/downloads/npu-dronemap-dataset/ which are referenced in the README can no longer be downloaded?

For example, clicking on https://pan.baidu.com/s/1gfwzcpt#path=%252FDataset%252FNPU%252FDroneMap%252Fphantom3-centralPark shows:

Thought I'd report it. 👍

Hello @zdzhaoyong
Thank for your great contribution. I want to ask you a question.
How to Chineseize the qviz display window？ such as setObjectName("Messenger Topics") , setObjectName("Frame Visualizer") and so on.

有没有关于Svar模块的代码基本介绍？这部分代码看的有点迷糊。

另外，你们PIL repo中的README.md链接打不开，不清楚是不是网速的问题。

感觉GSLAM是不错的框架！

/home/chen/GSLAM-master/GSLAM/plugins/datasets/DatasetTUMRGBD.cpp:94:37: error: conversion from ‘cv::Mat’ to non-scalar type ‘GSLAM::GImage’ requested

Hi I am trying to compile and install on ubuntu 20.10 and I am recieving the following error:

[ 10%] Building CXX object CMakeFiles/gslamDB_tummono.dir/GSLAM/plugins/datasets/DatasetTUMMono.cpp.o
/home/kevin/slam/GSLAM/GSLAM/plugins/datasets/DatasetTUMMono.cpp: In member function ‘bool DatasetTUMMono::loadVignette(const string&)’:
/home/kevin/slam/GSLAM/GSLAM/plugins/datasets/DatasetTUMMono.cpp:274:44: error: ‘CV_LOAD_IMAGE_GRAYSCALE’ was not declared in this scope
  274 |         cv::Mat vigMat=cv::imread(vigFile, CV_LOAD_IMAGE_GRAYSCALE);
      |                                            ^~~~~~~~~~~~~~~~~~~~~~~
/home/kevin/slam/GSLAM/GSLAM/plugins/datasets/DatasetTUMMono.cpp:277:23: error: no match for ‘operator=’ (operand types are ‘GSLAM::GImage’ and ‘cv::Mat’)
  277 |             _vignette=vigMat;
      |                       ^~~~~~
In file included from /home/kevin/slam/GSLAM/./GSLAM/core/GSLAM.h:66,
                 from /home/kevin/slam/GSLAM/GSLAM/plugins/datasets/DatasetTUMMono.cpp:5:
/home/kevin/slam/GSLAM/./GSLAM/core/GImage.h:215:13: note: candidate: ‘GSLAM::GImage& GSLAM::GImage::operator=(const GSLAM::GImage&)’
  215 |     GImage& operator=(const GImage& rhs)
      |             ^~~~~~~~
/home/kevin/slam/GSLAM/./GSLAM/core/GImage.h:215:37: note:   no known conversion for argument 1 from ‘cv::Mat’ to ‘const GSLAM::GImage&’
  215 |     GImage& operator=(const GImage& rhs)
      |                       ~~~~~~~~~~~~~~^~~
make[2]: *** [CMakeFiles/gslamDB_tummono.dir/build.make:63: CMakeFiles/gslamDB_tummono.dir/GSLAM/plugins/datasets/DatasetTUMMono.cpp.o] Error 1
make[2]: Leaving directory '/home/kevin/slam/GSLAM/build'
make[1]: *** [CMakeFiles/Makefile2:196: CMakeFiles/gslamDB_tummono.dir/all] Error 2
make[1]: Leaving directory '/home/kevin/slam/GSLAM/build'
make: *** [Makefile:130: all] Error 2

Any idea how I can fix this?

 LOG(ERROR)<<"Origin:"<<curGPS.get_rotation()
                     <<",Noise:"<<gps_noise.get_rotation()
                     <<",After:"<<noised_gps.get_rotation();
            LOG(ERROR)<<"Origin:"<<lnorigin
                     <<",Quat:"<<SO3::exp(lnorigin)
                     <<",After:"<<SO3::exp(lnorigin).ln();

Output as follows:

testOptimizer.cpp:420 Origin:0 0 -1 -1.71738e-16,Noise:-0.00122998 0.000986742 0.000211876 0.999999,After:-0.000782921 0.000628094 0.999999 -0.000212667
testOptimizer.cpp:423 Origin:-0.00245995 0.00197348 3.14202,Quat:-0.000782921 0.000628094 0.999999 -0.000212667,After:0.00245929 -0.00197295 -3.14117

The Origin and After should be equal!

GSLAM is loading and processing frames but not showing it in panel in GUI.
It just prints the following details:
Play thread stoped.

------------------------------------ Timer report ------------------------------------
FUNCTION #CALLS MIN.T MEAN.T MAX.T TOTAL

Dataset::grabFrame 407 1.8us 80.2ms 354.1ms 32.6 s
Main 1 738.5 s 738.5 s 738.5 s 738.5 s
--------------------------------- End of Timer report ---------------------------------

And not trajectory file is saved as required by Map2DFusion