jbehley / SuMa

Surfel-based Mapping for 3d Laser Range Data (SuMa)
MIT License
543 stars 166 forks source link

The visualizer crashed when click start button, something wrong with opengl #17

Closed wwtinwhu closed 5 years ago

wwtinwhu commented 5 years ago

Hi Behley, Could you help me with this problem? Thank you very much! Following is what in the terminal when I open visualizer and open one bin file and last click the start button:

./visualizer OpenGL Context Version 4.3 core profile GLEW initialized. OpenGL context version: 4.3 OpenGL vendor string : nouveau OpenGL renderer string: NV117 Extracting surfel maps partially. Performing frame-to-model matching. Gtk-Message: 15:12:49.587: GtkDialog mapped without a transient parent. This is discouraged. calibration filename: /media/weitongwu/Elements/Data/kitti/00/calib.txt...loaded. ground truth filename: /media/weitongwu/Elements/Data/poses/00.txt 4541 poses read. Performing frame-to-model matching. OpenGL Error: invalid value(1281) In: /home/weitongwu/suma_ws/src/glow/src/glow/GlFramebuffer.cpp on Line: 25 terminate called after throwing an instance of 'std::runtime_error' what(): OpenGL error detected. 已放弃 (核心已转储)

jbehley commented 5 years ago

It seems like that the nouveau driver does not implement all features that I use. Since Nvidia provides more then the core profile functionality with it's driver it happens sometimes that I use more than the required parts of the core profile. Thus, it is very hard for me to really test that I use only core parts of the OpenGL specification.

The easiest way to resolve this problem is to just use the Nvidia driver (if possible). If this is not possible, you could run the program with "gdb" and just provide me with the stack trace to find out where this problem happens.

wwtinwhu commented 5 years ago

@jbehley I am very sorry to reply you so late. I will try to use the Nvidia driver later and then if work I will close this issue, thank you very much!

LiShuaixin commented 5 years ago

The issue was caused by the wrong version of Gtsam. I met the same issue and I debugged the code to find out where the problem happens. The back trace showed the issue was occurred when calling gtsam function. Now, the only thing you need to do is recompiling gtsam of version 4.0.0-alpha2 and then recompiling SuMa

wwtinwhu commented 5 years ago

The issue was caused by the wrong version of Gtsam. I met the same issue and I debugged the code to find out where the problem happens. The back trace showed the issue was occurred when calling gtsam function. Now, the only thing you need to do is recompiling gtsam of version 4.0.0-alpha2 and then recompiling SuMa

@LiShuaixin Thanks for your advice. Recently, I am busy with something else. Then, I will try your method. Best Regards!

jbehley commented 5 years ago

@LiShuaixin Also thanks from me and I will then close the issue.

If the problem persists, please re-open the issue again.

zorosmith commented 4 years ago

Hello @LiShuaixin . You solve my problem! Could you tell me how did you figure out this problem?

xdtzzz commented 2 years ago

hi,@zorosmith,what version number is your computer, i use 4.0.0-alpha2, but it still happen this problem.what should i do?

jbehley commented 2 years ago

I have it running with ubuntu 20.04, but it also compile and worked with Ubuntu 18.04. important is to compile gtsam with "use_system_eigen" option enabled. Usually, you will get some error in some matrix computation if you did not compile with the system's Eigen version.

xdtzzz commented 2 years ago

thanks reply so quickly, I wonder what eigen version corresponding gtsam4.0.0-alpha2? In my computer, I can run SuMa++ successfully, but I can't run SuMa