- in fact, put ALL the code of debug in arucocontext
- expose debug info in arucocontext
- expose all the hardcoded parameters from AR.Detect in the debug.html
  - init gui in debug.html
  
  

# Fix pose estimation
- epnp as way to fix pose estimation issue
- https://infoscience.epfl.ch/record/160138/files/top.pdf?version=1
- standalone cpp implementation - https://github.com/artivis/epnp/blob/master/src/main.cpp

# Performances
- jsaruco use a kernel size of 2 in adaptative thresholding
  - could i reduce the resolution of the source image and use a kernel size of 1 ?
  - it would produce more fps. what the difference would be ? create errors ?
- jsaruco - adaptiveThreshold is doing it on ALL bytes - so all channel ??? check if it is correct
  - it use blackwhite image - it only needs 1 channel - 8 bits is already a lot to store blackwhite
  - this mean 4 times more work than needed
  - NOTES: unclear this is true - grayscale is packing it all in 1 channel. check it out ?
- in posit1: image difference is computed by a manathan distance. not a euclidian distance... how come ?
  - this seems to intrduce an error without need (sqrt is nothing in js)
  - ```imageDifference += Math.abs(sopImagePoints[i].x - oldSopImagePoints[i].x);```
  - ```imageDifference += Math.abs(sopImagePoints[i].y - oldSopImagePoints[i].y);```

# How to include it in AR.js
- see how to include it in ar.js
  - so basically a context and a controls
  - the source can remain the same without trouble
- how to handle all the options between the various backend

- how to handle the development ? 
  - do you copy the files ? 
  - do i modify the master files

- seems to have only a few modifications
  -