Lighting variation is commonly handled by methods invariant to additive and multiplicative changes in image intensity. It has been demonstrated that comparing images using the direction of the gradient can produce broader insensitivity to changes in lighting conditions, even for 3D scenes. We analyze two common approaches to image comparison that are invariant, normalized correlation using small correlation windows, and comparison based on a large set of oriented difference of Gaussian filters. We show analytically that these methods calculate a monotonic (cosine) function of the gradient direction difference and hence are equivalent to the direction of gradient method. Our analysis is supported with experiments on both synthetic and real scenes.
On the equivalence of common approaches to lighting insensitive recognition
Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 ; 2 ; 1721-1726 Vol. 2
2005-01-01
188272 byte
Conference paper
Electronic Resource
English
On the Equivalence of Common Approaches to Lighting Insensitive Recognition
British Library Conference Proceedings | 2005
|Comparing and combining lighting insensitive approaches for face recognition
British Library Online Contents | 2010
|Illumination insensitive recognition using eigenspaces
British Library Online Contents | 2004
|Viewpoint Insensitive Action Recognition Using Envelop Shape
Springer Verlag | 2007
|A Study on View-Insensitive Gait Recognition
British Library Conference Proceedings | 2005
|