Appearance Decomposition for Image-based Reconstruction
Speaker: Todd Zickler , Harvard University
Date: May 3 2005
Time: 2:45PM to 3:45PM
Location: G463 (Star)
Host: Greg Shakhnarovich, CSAIL
Contact: Greg Shakhnarovich, xx3-8170, gregory@csail
Relevant URL: PLEASE NOTE the unusual DAY (Tuesday instead of Wednesday) and LOCATION (Star instead of Kiva).
ABSTRACT:
Image-based reconstruction systems are designed to accurately recover
the three-dimensional shape of a scene from its two-dimensional
images. Applications include visual inspection, reverse engineering,
digital object archival, enrollment for 3D face/object recognition
systems, and virtual/augmented reality. The reconstruction problem is
ill-posed because images do not generally provide direct access to 3D
shape. Instead, shape information is coupled with additional factors:
illumination, pose and reflectance.
Most existing reconstruction methods obtain 3D shape by making
assumptions about reflectance. For example, it is often assumed that
surfaces are perfectly matte or diffuse (i.e., Lambertian.) The
applicability of these methods is limited because the accuracy of the
recovered shape can be compromised when the underlying assumptions are
violated.
In this talk I present two techniques for recovering 3D shape without
making restrictive assumptions about surface reflectance. Both
methods are based on the notion of appearance decomposition: by
uncoupling some of the factors that determine an image (shape,
reflectance, illumination and pose), we obtain more direct access to
shape information, greatly simplifying the reconstruction problem. I
demonstrate how these methods can be applied to a much broader class
of surfaces, and discuss their applications.
For more information, see
http://www.eecs.harvard.edu/~zickler/research.html
BIO:
Todd Zickler received his B. Eng. in Electrical Engineering from
McGill University (Montreal, Canada) in 1997. He completed his
M.S. and Ph.D. in Electrical Engineering at Yale University in 2001
and 2004, respectively. In 2004, he joined the Division of
Engineering and Applied Sciences at Harvard University as an Assistant
Professor of Electrical Engineering. Zickler's current research focus
is in the field of computer vision---working to build systems that
visually understand and interact with their environment. He has
developed new methods for recovering the shape and material properties
of scenes from their images, and is generally interested in acquiring,
representing and applying visual appearance information.
See other events that are part of MIT Machine Vision Colloquium 2004/2005
See other events happening in May 2005
