TOD detection crashes when training has been done at 1280x1024 pixels

[jordi-pages]

This crash occurs when the objects have been trained with a Kinect at 1280x1024 pixels. Then, when running the detection as follows:

rosrun object_recognition_core detection -c `rospack find object_recognition_tod`/conf/detection.ros.ork --visualize

the following crash appears:

OpenCV Error: Assertion failed (func != 0) in transpose, file /home/hudson/tmp/opencv/opencv-2.4.6.1/modules/core/src/matrix.cpp, line 2023
Traceback (most recent call last):
  File "/home/jordi/branches_svn/hydro_migration/ork_ws/src/object_recognition_core/apps/detection", line 24, in <module>
    run_plasm(args, plasm)
  File "/opt/ros/hydro/lib/python2.7/dist-packages/ecto/opts.py", line 85, in run_plasm
    sched.execute(options.niter)
ecto.CellException:            exception_type  CellException
                cell_name  pipeline1
                cell_type  ecto::py::BlackBox
            function_name  process
                     type  std::runtime_error
                     what             exception_type  CellException
                cell_name  Matcher
            function_name  process
                     type  cv::Exception
                     what  /home/hudson/tmp/opencv/opencv-2.4.6.1/modules/core/src/matrix.cpp:2023: error: (-215) func != 0 in function transpose

                     when  While triggering param change callbacks

This crash happens independently of the resolution at which the Kinect is operating during the detection, i.e. 640x480 or 1280x1024. That is why I guess it is a problem managing the training data stored in the database.

[kelayamatoz]

Hi jordi,

I am getting exactly the same problem as you are having here. Wondering that if you were able to figure out a solution? Many thanks!

Best,

Tian