Rendering - Image-Based Rendering

Model-Based Recognition

Project Overview

Model-based recognition seeks to match a parameterized model with one or more images that are known prior to depicting an instance of the model; the model may be as simple as a surface of revolution or as complex as a human face [DU94]. The latter example is clearly applicable to telecollaboration, where human faces are the dominant features of the scene.


Center Sites

Brown, Caltech, Cornell

Lead Researchers

Jim Arvo
John H. Hughes
Ken Torrance

Bibliographic References

[DU94]K. Delibasis and P.E. Undrill, "Anatomical object recognition using deformable geometric models", Image and Vision Computing, 12(7), 1994, 423-433.

Rendering Bibliography

Full Research Bibliography

Web References

information extraction Inverse and Interpolation Methods for Image Synthesis

Rendering Overview
Home Research Outreach Televideo Admin Education