Ethzasl_ptam camera model [closed]

asked 2012-11-27 03:53:03 -0600

BeLioN gravatar image

updated 2014-01-28 17:14:24 -0600

ngrennan gravatar image


I just started working with the ethzasl_ptam stack , and the first thing that got my attention was that the camera model used is really simple.

It just relies on 5 parameters: fx, fy, cx, cy and s.

This means that only 1 parameter is used to model the distortion of the lens. In contrast, OpenCV can use 4, 5 or 8 parameters only to model the distortion.

What can be the impact of using such a simple model? Do you think that it would be possible to improve the matching rate and the accuracy by improving the camera model used?


edit retag flag offensive reopen merge delete

Closed for the following reason question is not relevant or outdated by tfoote
close date 2015-08-31 20:14:35.244731