@okreylos really nice, next step undistorting and on the search for 3 or 4 point collinearity ?pic.twitter.com/XpD6V8lAbe
After accounting for a bug in its firmware, this is what the DK2's tracking camera *really* sees:pic.twitter.com/fnKUhsL40d
-
-
-
@gerwindehaan Now it gets hard. We have the 3D model and 2D observations, meaning we can directly calculate intrinsics and lens distortion. -
@okreylos crowdsourced calibration FTW. Anyone modelled 3D LED pos yet? Reminds me of http://homepages.cwi.nl/~robertl/papers/2003/ieeevr03/paper.pdf … and later#PSTechNL -
@gerwindehaan Thanks for the paper! We have 3D LED positions from firmware, and those you indicated aren't actually collinear in 3D. :(
End of conversation
New conversation -
-
-
.
@okreylos What was the nature of the bug? -
@Eric_Neuman It advertised as 376x480 YUYV instead of 752x480 Y8. Wrong resolution and pixel format, but same frame data size.
End of conversation
New conversation -
-
@okreylos trying to get this working in opencv under windows/DirectShow. any more info about your unpacker? is magic in ImageExtractorBA81?Thanks. Twitter will use this to make your timeline better. Undo
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.