Welcome new user! You can search existing questions and answers without registering, but please register to post new questions and receive answers. Note that due to large amounts of spam attempts, your first three posts will be manually moderated, so please be patient.
Because of un-manageable amounts of spam despite our use of CAPTCHAs, email authorization, and other tools, we have discontinued this forum (see the 700k+ registered users with validated email addresses at right?). Please email us any questions or post bug reports and feature requests on GitHub at https://github.com/jevois -- The content below remains available for future reference.
Welcome to JeVois Tech Zone, where you can ask questions and receive answers from other members of the community.

how to un-distort 90deg fov image with gpu?

0 votes


I am toying around with the AR0135 
it comes with a default 3.6mm lens (88deg)

I am thinking of using it to observe markers on a flat ground for indoor localization.
I kinda like this near 90 field of view. but i do observe quite some distortion. Look on this preview with a white flat A4 paper is not a rectangle at all (not too mention that the sides of the image seem darkish and fuzzy)

i am pretty sure i have read -- maybe on the forum => http://jevois.org/qa/index.php?qa=153 -- that jevois comes with software to un-distort the image with its gpu somehow. i'd like to test that. but i am unsure of the vocabulary to use in order to search in the docs and no default module seems to be doing that. 

what's the best way to get started? 


asked Mar 12, 2019 in User questions by fourchette (580 points)

1 Answer

0 votes
yes, following that post that you found, we wrote this module:


which includes a barrel distortion shader (called Dewarp).

source here: http://jevois.org/basedoc/jvpkg_2modules_2JeVois_2DemoGPU_2DemoGPU_8C_source.html

Now, this is C++ only and you should use this component in your module: http://jevois.org/basedoc/classFilterGPU.html

Another approach is to use the CPU and OpenCV in python. A quick search on the web returns lots of results, for example:

answered Mar 13, 2019 by JeVois (46,580 points)