Floating point error when running mask_rcnn example code from learnopencv C++

1

Hi All:

I compiled and installed opencv into a local directory, using the Ubuntu 18.04 installation script…

installOpenCV-3-on-Ubuntu-18-04.sh

I compiled the example C++ code from here successfully.

When I run I get floating point errors shortly after a QT window pops up.

./learnopencv/Mask-RCNN/build$ ./mask_rcnn --image=…/cars.jpg

qt5ct: using qt5ct plugin
Floating point exception (core dumped)

I’m not sure how to solve this problem.

2

please explain what is “the Ubuntu 18.04 installation script”

you likely followed instructions that are outdated or flawed.

further, use a debugger to determine where in your code this issue happens.

3

Hi

I used this script.

This is the first use of openCV in C++ after install. Is there a way to see if my opencv installation is sane for deep learning before trying the example code?

4

After some debugging I learned that there was a problem with inputs related to color.txt that I resolved.
Thanks! You can close this as solved.