Home

Bits of cotton

Welcome to Goki check wiki!

Goki check is vectorized graphics matching program, planning to be implemented.
So good prepared 3D models leads us to get context from graphics.
It is planned that fisheye.py transformation then 3D to 3D matching to be implemented,
but I have no time to implement, it'll be late.

  • Web demo is @ konbu.sakura.ne.jp .
  • I want these program be used widely without restricting parent license, the license is public domain now, and Goki check and small tools in here are no warranty.

  • [Basic idea] : basic ideas for matching models.

Enlarge before vectorize:
[enlarge.py] : Enlarge images before edge detecting.
[enlarge-ds.py] : enlarge.py for differential space.

Vectorize:
(Photonics has prior works) :
[edgedetect.py] : Not using curvelet, but using enlarge.py algorithm.
[edgecollect.py] : With edgedetect.py, absolute value.

(If images have a correct colour) :
[fisheye.py]: build bump mapping plausible image from 2D image without depth camera. (In version 0.06, it complement lack of informations on color flat places but is adoptic method.)
fisheye.py before 0.06.1 have glitches, after 0.07, we can get plausible ones.
fisheye.py before 0.07 have non statistics source (usually gains nonsense regions.), after 0.08 we can get statistical plausible ones.
fisheye.py 0.09 has 2x width with horizontal and vertical direction.
* Because of adoptic colour ommission, fisheye.py 0.09 often have striped image.

Project Admins:


Related

Wiki: Basic idea
Wiki: edgecollect.py
Wiki: edgedetect.py
Wiki: enlarge.py
Wiki: fisheye.py

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks