15 articles 

Workshop in conjunction with ICCV'07, Rio de Janeiro, Brazil, October 14-21, 2007

Scope:

The way an image looks like depends on many factors, including geometry, illumination and reflectance properties of the objects. For the transparent or translucent objects, or for the objects composed by multiple coatings, the factors are even more numerous (refraction, subsurface scattering,...). The laws combining these components are very diverse and complex. This complexity makes computer vision tasks even more difficult and practically causes the failure of methods based on too simple models.
A typical example could be the troubles caused by the specularities in the stereo-vision problem; proposed methods usually assume that the scene in perfectly diffuse. Feature tracking/matching is another example since the photometric appearance of the objects can change when they/the camera move/es.
From the theoretical as well as from the computational point of view, a better understanding and handling of these factors and of their combinations should allow to be robust to the photometric effects. In fact this allows us to go beyond: it allows not only to overcome the inconveniences problems they involve but it can also be an information/constraints source which can be practically exploited in computer vision tasks. We can think for example about the shading and shadow information.

Best Papers:

MERL Keynote Lecture:

Organizing Committee:

Program Committee:

Sponsors: