Yeah you are right good catch, my correction is wrong. There is indeed an additional typo on the tl.y >= im_grey.cols check that should be tl.y >= im_grey.rows But I'm still not sure the -1 is enough because in the next lines there is a im_grey.at<uchar>(tl.y+1,tl.x+1) which will be out of bounds because if tl.y == im_grey.rows -1 then tl.y + 1 = im_greys.rows. Thanks</uchar> Le jeu. 2 juin 2022 à 18:34, John Baldwin baldwj@users.sourceforge.net a écrit : Ah, I was ignoring this error putting the...
Ah, I was ignoring this error putting the tracker in a try catch block. I finally decided to look into it this week. Found a solution and came here to put in a bug report and found that you beet me to the punch :) Anyways I'm curious if -2 is the proper solution. for myself I was changing the check for tl.y to be on the rows rather than the columns. if(tl.y>=im_grey.cols)tl.y=im_grey.rows-1 to if(tl.y>=im_grey.rows)tl.y=im_grey.rows-1 This fix makes sense to me. My rows and columns are different...
Crash in aruco_dcf 2
Crash in aruco_dcf
I was also getting the same error. I updated my opencv to version 4.5.4 and tried again and it was built successfully.
Hi ! Your CPU/Computer mostly affect the processing speed. The pose accuracy is mostly affected by the camera hardware, in particular its resolution (the bigger the better) in comparison to its field of view, as well as the marker distance from the camera (the further away the marker, the more imprecision increases). (And of course the algorithms, but let's consider the ArUco one here). For the choice of camera/CPU, it depends on your goal. The bigger the resolution, the more processing is needed,...
Hi ! Your CPU/Computer mostly affect the processing speed. The pose accuracy is mostly affected by the camera hardware, in particular its resolution (the bigger the better) related to its field of view, as well as the marker distance from the camera (the further away the marker, the more imprecision increases). (And of course the algorithms, but let's consider the ArUco one here). For the choice of camera/CPU, it depends on your goal. The bigger the resolution, the more processing is needed, the...
Hi ! Your CPU/Computer mostly affect the processing speed. The pose accuracy is mostly affected by the camera hardware, in particular its resolution (the bigger the better) related to its field of view, as well as the marker distance from the camera (the further away the marker, the more imprecision increases). (Anf of course the algorithms, but let's consider the ArUco one here). For the choice of camera/CPU, it depends on your goal. The bigger the resolution, the more processing is needed, the...
Hi ! Your CPU/Computer mostly affect the processing speed. The pose accuracy is mostly affected by the camera hardware, in particular its resolution (the bigger the better) and its field of view (the smaller the better), as well as the marker distance from the camera (the further away the marker, the more imprecision increases). For the choice of camera/CPU, it depends on your goal. The bigger the resolution, the more processing is needed, the lower your FPS will be. If you require 30 FPS with a...
Hi ! Your CPU/Computer mostly affect the processing speed. The pose accuracy is mostly affected by the camera hardware, in particular its resolution (the bigger the better) and its field of view (the smaller the better), as well as the marker distance from the camera (the further away the marker, the more imprecision increases). For the choice of camera/CPU, it depends on your goal. The bigger the resolution, the more processing is needed, the lower your FPS will be. If you require 30 FPS with a...
Unimplemented interface in MarkerDetector
Aruco-3.1.12+Ubuntu 18.04
Effects of hardware on pose estimation accurcy
DM_VIDEO_FAST cornerUpsample-Bug and Workaround
Hello Robert, It seems like my bug in DM_VIDEO_FAST originates in the "Autosize" feature. Changing the "params.ts" value altering the MinMarkerSize that is used by the implementation of the markerdetector. This breaks the Resize or the ImagePyramid feature resulting in an error in the ID detection even though the contours are detected correctly (as far as i can tell). If you are having troubles with DM_FAST the only change from DM_NORMAL is the ThresholdMethod. markerdetector.cpp line 384-394: "...
Hello Robert, i also want to use the Library for the DM_FAST and the DM_VIDEO_FAST modes. DM_FAST works fine for me on my completely new installed Ubuntu 20.04 with OpenCV 3.4.13 and ArUco 3.1.12. However i also noticed DM_VIDEO_FAST not functioning correctly see picture attached. I tried it with and without a camera.yml and in the YT-Videos you can see Rafael Muñoz Salinas using it both ways and it functions correctly. In the Screenshot you can see the purple "show all"-indicators moving in the...
Hello Robert, i also want to use the Library for the DM_FAST and the DM_VIDEO_FAST modes. DM_FAST works fine for me on my completely new installed Ubuntu 20.04 with OpenCV 3.4.13 and ArUco 3.1.12. However i also noticed DM_VIDEO_FAST not functioning correctly see picture attached. I tried it with and without a camera.yml and in the YT-Videos you can see Rafael Muñoz Salinas using it both ways and it functions correctly. In the Screenshot you can see the purple "show all"-indicators moving in the...
DM_FAST and DM_VIDEO_FAST consistently throw exceptions on detect()
is it possible to recognize 5x5 dictinoary markers with aruco_simple?
[CMake] v3 can't be build with add_subdirectory anymore (should use PROJECT_SOURCE_DIR)
[Feature Request] Ground Plane
Note: Error persists in Aruco 3.1.11
CMakeOutput.log in addition: The system is: Darwin - 18.7.0 - x86_64 Compiling the CXX compiler identification source file "CMakeCXXCompilerId.cpp" succeeded. Compiler: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/c++ Build flags: Id flags: The output was: 0 Compilation of the CXX compiler identification source "CMakeCXXCompilerId.cpp" produced "a.out" The CXX compiler identification is AppleClang, found in "/Users/frothe/Source/aruco-3.1.2/build/CMakeFiles/3.17.0/CompilerIdCXX/a.out"...
CMake Error when building with flag INSTALLl_DOC on
Aruco Visual Studio build errors
marker map test feedback
Yes, it is. I had calibrated the camera with aruco_calibration.exe several times and generated the .yml file. I got an error around 2 . However, to answer you I decided to repeat the whole process, and I got in calibrate an error of 0.6. And, with that new file, it works perfectly! Sorry for the inconvenience, and thanks for responding! :-) You can close the 'issue'. JC PS: I have suspicions about what could have happened... For some reason aruco_calibration.exe doesn't accept my video stream. So...
Yes, it is. I had calibrated the camera with aruco_calibration.exe several times and generated the .yml file. I got an error around 2 . However, to answer you I decided to repeat the whole process, and I got in calibrate an error of 0.6. And, with that new file, it works perfectly! Sorry for the inconvenience, and thanks for responding! JC PS: I have suspicions about what could have happened... For some reason aruco_calibration.exe doesn't accept my video stream. So I have to generate a video, convert...
Did u calibrate the camera? El 18/2/20 a las 14:56, JCarlos escribió: [bugs:#51] https://sourceforge.net/p/aruco/bugs/51/ Bug in aruco::CvDrawingUtils::draw3dAxis and aruco::CvDrawingUtils::draw3dCube (aruco-3.1.11) Status: open Group: v1.0 (example) Created: Tue Feb 18, 2020 01:56 PM UTC by JCarlos Last Updated: Tue Feb 18, 2020 01:56 PM UTC Owner: nobody Attachments: 2020-02-14_134439.png https://sourceforge.net/p/aruco/bugs/51/attachment/2020-02-14_134439.png (578.3 kB; image/png) 2020-02-14_134831.png...
Bug in aruco::CvDrawingUtils::draw3dAxis and aruco::CvDrawingUtils::draw3dCube (aruco-3.1.11)
Dictionary is a set of markers. Each library developer has proposed a set of valid markers they detect. I call this a Dictionary of markers. Aruco can detect markers from many others libraries El 17 ene. 2020 4:13, pwd <pwd@users.sourceforge.net> escribió: [bugs:#50] the 'ditionaries' in the file README Status: open Group: v1.0 (example) Created: Fri Jan 17, 2020 03:13 AM UTC by pwd Last Updated: Fri Jan 17, 2020 03:13 AM UTC Owner: nobody Does the 'ditionaries' mean dictionaries in the file README...
the 'ditionaries' in the file README
3.1.8 build errors with Visual Studio 2017 on Windows 10 x64
@cv-hobbyist i have the same problem, anyone have idea。
The project is OK now,thanks for your work.
Ive uploaded a new version. Let me know if the problem persists. Thanks El 7/11/19 a las 4:42, Fan PI escribió: Sorry for bothering you again.Does the "-s" parameter represent the side length of the outermost maker?If I set the value like that, the 3d cube result picture is totally wrong,but if I set a far more bigger value,the 3d cube result picture is right. *[bugs:#48] https://sourceforge.net/p/aruco/bugs/48/ Fractal Markers project does not use the parameter of markersize * Status: open Group:...
My phd student has discovered a bug. He is working on that. Thanks for testing. Ill keep u posted El 7 nov. 2019 4:42, Fan PI <pivapi@users.sourceforge.net> escribió:Sorry for bothering you again.Does the "-s" parameter represent the side length of the outermost maker?If I set the value like that, the 3d cube result picture is totally wrong,but if I set a far more bigger value,the 3d cube result picture is right. [bugs:#48] Fractal Markers project does not use the parameter of markersize Status:...
Sorry for bothering you again.Does the "-s" parameter represent the side length of the outermost maker?If I set the value like that, the 3d cube result picture is totally wrong,but if I set a far more bigger value,the 3d cube result picture is right.
By the way,the name of zip file may be named "aruco3.1.7.zip",not "aruco3.7.1.zip",thank you again.
I really appreciate your help.
Problem solved. Thank for your help. Please check the new version 3.1.7. El 6/11/19 a las 5:00, Fan PI escribió: *[bugs:#48] https://sourceforge.net/p/aruco/bugs/48/ Fractal Markers project does not use the parameter of markersize * Status: open Group: v1.0 (example) Labels: Fractal Created: Wed Nov 06, 2019 04:00 AM UTC by Fan PI Last Updated: Wed Nov 06, 2019 04:00 AM UTC Owner: AVA Group Hiya: The Fractal Markers project is wonderful,but I face a problem,in the file of "fractalposetracker.cpp",the...
Thanks a lot, ill check it ritgh now El 6/11/19 a las 5:00, Fan PI escribió: *[bugs:#48] https://sourceforge.net/p/aruco/bugs/48/ Fractal Markers project does not use the parameter of markersize * Status: open Group: v1.0 (example) Labels: Fractal Created: Wed Nov 06, 2019 04:00 AM UTC by Fan PI Last Updated: Wed Nov 06, 2019 04:00 AM UTC Owner: AVA Group Hiya: The Fractal Markers project is wonderful,but I face a problem,in the file of "fractalposetracker.cpp",the function of "FractalPoseTracker::setParams"...
Fractal Markers project does not use the parameter of markersize
Does the change to the GPLv3 license affect the licensing status for the aruco module in OpenCV (https://github.com/opencv/opencv_contrib/tree/master/modules/aruco)? What was the first version of this library to have a GPLv3 license?
Idea Binary Scoring and OI TOKEN
Current 3.1.2 release is GPLv3 with obfuscated code. Is it even compatible with the GPLv3 license (SO post)? They have removed old source code that were under a permissive license (BSD2?). All these moves show a nice spirit from them...
Image attached since it seems that I cannot edit the ticket...
Fractal marker is not correctly generated
I have the same problem, also I can't build marker_mapper 1.0.12 with aruco 3.1.2.
ippe.cpp computeTranslation error
Thanks a lot for your help. Im releasing the new version very soon.
3.1.1 windows build errors
network gstreamer rtsp argument or loopback causes exceptions
aruco_mip_36h12_dict (inter-markerdistnace)
3.1.0 missing source code?
opencv4 compile error
Thank you Rodrigues. Your sent paper and this last sent official documents really served well for me. Best regards, Hamid On Mon, Mar 11, 2019 at 10:09 AM Rodrigo rodriglez@users.sourceforge.net wrote: Here you´ve got some oficial documentation: https://docs.google.com/document/d/1QU9KoBtjSM2kF6ITOjQ76xqL7H0TEtXriJX5kwi9Kgc/edit [bugs:#39] https://sourceforge.net/p/aruco/bugs/39/ Generating a custom dictionary with a good inter-marker distance* Status: open Group: v1.0 (example) Created: Fri Mar...
Here you´ve got some oficial documentation: https://docs.google.com/document/d/1QU9KoBtjSM2kF6ITOjQ76xqL7H0TEtXriJX5kwi9Kgc/edit
Thank you a lot RODRIGO... great article, By the way, I have one question of you. Do you have any information about the ArUco marker calculation. For example I would like to create a marker configuration like a octagoan shape. Now, how Can I select markers such that errors rates like the inter-marker distances becomes less. On Fri, Mar 8, 2019 at 4:51 PM Rodrigo rodriglez@users.sourceforge.net wrote: [bugs:#39] https://sourceforge.net/p/aruco/bugs/39/ Generating a custom dictionary with a good inter-marker...
Generating a custom dictionary with a good inter-marker distance
Hello David, Thank you for answering. I think my main problem is that I am not adept in CMake. Can you show me or write it as line how should I show my OpenCV directory in CMake. By the way Can I have your CMake as a sample to take help? Best Regards, Hamid
Hello, I had no issue to install Aruco on W10. I think you just need to include the OpenCV directory inside your path
ArUco 3.0.13 windows 10 installation
OpenCV ArUco Dictionaries
Hello, Hello David, I solved the problem,. I removed the assert at the begining of the "impl__aruco_getRTMatrix" in posetracker.cpp. And it worked,. Thanks a lot for your comments.
Yes to solve the issue you need to modify the source code and rebuild the library. If the problem still persists, I'm not sure what could be the cause. Do you have a different programming environment ? That may be the cause. Initially, to find the error, I observed that the output of tvecs and rvecs using std::cout was correct, but not once set as floats. Do you observe a similar behaviour ?
Dear friends, I tried both the above method but erroe is still exists??!!!!