Markerless Tracking/Related Work

This page sumarizes related work for the research project Markerless Tracking.

Recognition
[http://mi.informatik.uni-siegen.de/ Blanz et. al.] showed how to use computer generated images of faces to recognize real ones. This work is based on a morphable face model which was presented before.

Not jet reviewd:

Tracking
Motion Capturing could greatly improve with analysis-by-synthesis as stated by Moeslund and Granum.

Not jet reviewd:

Theory
Not jet reviewd:

Commercial
Organic Motion's solution to Motion Capturing was presented at SIGGRAPH 2007 (demo video) and is the first commercial available product to track all bodyparts including bones, geometry and textures. Their website claims that "Tschesnok created a thinking system which looks at people in a manner very similar to the way the brain process human vision." . Obviously it is not quite clear how this works. Hopefully their pending patent will be available soon.

Interdisciplinary
As briefly described in neuro-scientists found evidence that the nervous system use probabilistic population codecs.