Inverse Rendering with a Morphable Model: A Multilinear Approach
Oswald Aldrian and Will Smith
In: BMVC 2011, 29 Aug - 2 Sep 2011, Dundee, UK.
Well known results in inverse rendering show that recovery of unconstrained illumination, texture and reﬂectance properties from a single image is ill-posed. On the other
hand, in the domain of faces linear statistical models have been shown to efﬁciently
characterise variations in face shape and texture. In this paper we show how the inverse
rendering process can be constrained using a morphable model of face shape and texture. Starting with a shape estimate recovered using the statistical shape model, we show
that the image formation process leads to a system of equations which is multilinear in
the unknowns. We are able to estimate diffuse texture, specular reﬂectance properties,
the illumination environment and camera properties from a single image. Our approach
uses relaxed assumptions and offers improved performance in comparison to the current
state-of-the-art morphable model ﬁtting algorithms.
|EPrint Type:||Conference or Workshop Item (Paper)|
|Project Keyword:||Project Keyword UNSPECIFIED|
|Deposited By:||Will Smith|
|Deposited On:||12 February 2012|