0

I've been looking into correcting images taken with a fisheye lens and was wondering where OpenCV got their algorithm from? Was it based on a research paper someone published or is it an algorithm developed internally? I have tried looking around on their website for a bibliography or something but I haven't found any (although I could just not be seeing it). This also makes me wonder then is it consistently implementing algorithms developed in academia or constantly developing their own, or some kind of mix of the two?

  • Can you provide the code (function etc) you are referring to? – Eypros Jul 21 '20 at 21:28
  • no clue for OpenCV but fisheye corrections are well known from Wolfenstein times ... using simple geometry/trigonometry and similarity of triangles in combination with ray tracing – Spektre Jul 22 '20 at 06:12
  • @Eypros More so in relation to the detailed description section of [this documentation](https://docs.opencv.org/3.4/db/d58/group__calib3d__fisheye.html) – Puffleballz Jul 22 '20 at 13:03
  • @Spektre Then why have their been papers written in academia about how to correct them relatively recently (10ish years ago). Is it because of trying to get it to run efficiently on a computer in realtime versus doing all sorts of equations each time? – Puffleballz Jul 22 '20 at 13:05
  • @Puffleballz hard to say without reading them.... my guess is using HLSL/GLSL/CUDA/CL based techniques to improve speed ... – Spektre Jul 22 '20 at 15:17

0 Answers0