Documentation of SGBM Stereo is not exact (Bug #1937)


Added by Antonio Albiol almost 13 years ago. Updated over 11 years ago.


Status:Done Start date:2012-05-15
Priority:Normal Due date:
Assignee:Daniil Osokin % Done:

0%

Category:documentation
Target version:2.4.4
Affected version:branch '2.4' (2.4-dev) Operating System:Any
Difficulty: HW Platform:Any
Pull request:https://github.com/Itseez/opencv/pull/362

Description

In the Documentation of

StereoSGBM::StereoSGBM()

it says:

"speckleRange – Maximum disparity variation within each connected component. If you do speckle filtering, set the parameter to a positive value, multiple of 16. Normally, 16 or 32 is good enough."

In fact there is no need that this param is multiple of 16. May be in an older version it was necessary but now it is not.

The call to the relevant function in 2.4.0 is:
filterSpeckles(disp, (minDisparity - 1)*DISP_SCALE, speckleWindowSize, DISP_SCALE*speckleRange, buffer);
and I guess that DISP_SCALE is 16.

The unit of speckleRange, as it is in the code now, I think that it is "pixels of disparity", and may be this should be reflected in the documentation. Normal values, should be 1 or 2.


Associated revisions

Revision f9de98ec
Added by Daniil Osokin about 12 years ago

Fixed proposed values for speckleRange in StereoSGBM docs (bug #1937)

Revision f0049fa2
Added by Andrey Pavlenko over 11 years ago

Merge pull request #1937 from ilya-lavrenov:tapi_integral

History

Updated by Kirill Kornyakov almost 13 years ago

  • Category set to documentation

Updated by Vadim Pisarevsky over 12 years ago

  • Target version deleted ()

Updated by Kirill Kornyakov over 12 years ago

  • Target version set to Next Hackathon

Updated by Vadim Pisarevsky about 12 years ago

  • Target version deleted (Next Hackathon)

Updated by Daniil Osokin about 12 years ago

  • Affected version set to branch '2.4'
  • Assignee set to Daniil Osokin

Updated by Daniil Osokin about 12 years ago

  • Pull request set to https://github.com/Itseez/opencv/pull/362

Updated by Andrey Kamaev about 12 years ago

Fix is pushed to 2.4

  • Status changed from Open to Done

Updated by Kirill Kornyakov about 12 years ago

  • Target version set to 2.4.4

Updated by Shan Huang over 11 years ago

Please leave some warnings in documents, it takes me a week to reproduce experiment results in old opencv.

Updated by Sergei Nosov over 11 years ago

What warning do actually want to be in docs?

  • HW Platform set to Any
  • Operating System set to Any
  • Affected version changed from branch '2.4' to branch '2.4' (2.4-dev)

Also available in: Atom PDF