0

For a more complex project, I need to compute the approximate, relative distances of objects from two images (from stereo-cameras). Practically what this neat tutorial explains: https://chrisjmccormick.wordpress.com/2014/01/10/stereo-vision-tutorial-part-i/ and with a result like that enter image description here

Think I shouldn't be reinventing the wheel for this project and since speed is very important (realtime from two videostreams) I'm looking for a native library (preferably in C++ where the whole project is written in) for this task.

Does anyone have a suggestion? Open source would be greatest but not mandatory.

Huge thanks in advance!

DragonGamer
  • 834
  • 3
  • 9
  • 27

1 Answers1

1

try with LIBELAS library (Library for Efficient Large-scale Stereo Matching).

Best!

taihu
  • 72
  • 1
  • 7
  • Not sure you aren't here to advertise a project you might be involved with yourself, but that you anyways :) That does look promising yes! Also justs tumbled across some comparison with OpenCV which I would have tried myself.. This liberary you linked is said to be a lot faster. – DragonGamer Jul 07 '16 at 15:33
  • hahaha I would like to be part of LIBELAS project :) I am not sure if it is faster but it looks like it provide a denser disparity map than OpenCV. – taihu Sep 19 '16 at 17:29