In my spare time, I've recently been exploring some augmented reality projects. One challenge with existing systems is the disconnect between computer vision systems and 3D renderers. In particular, it would be really nice if it were easy (or at least possible) to add a camera distortion model to the renderer, so as to "redistort" a rendered model to a video.
Cheers! Kevin On Monday, May 19, 2014, Jason Grout <[email protected]> wrote: > On 5/19/14, 10:52, Jason Grout wrote: > >> (see http://sagecell.sagemath.org/?q=vtjadv for an example of a point >> tracking the mouse on a sphere) >> > > Sorry; the correct link for the example is http://sagecell.sagemath.org/? > q=yiarsu > > Thanks, > > Jason > >
