In Search of Illumination Invariants
نویسندگان
چکیده
We consider the problem of determining functions of an image of an object that are insensitive to illumination changes. We rst show that for an object with Lambertian re ectance there are no discriminative functions that are invariant to illumination. This result leads us to adopt a probabilistic approach in which we analytically determine a probability distribution for the image gradient as a function of the surface's geometry and re ectance. Our distribution reveals that the direction of the image gradient is insensitive to changes in illumination direction. We verify this empirically by constructing a distribution for the image gradient from more than 20 million samples of gradients in a database of 1,280 images of 20 inanimate objects taken under varying lighting condition. Using this distribution, we develop an illumination insensitive measure of image comparison and test it on the problem of face recognition.
منابع مشابه
A Unified Framework for Image-Derived Invariants
Image features and shape descriptors that capture the essential traits of an object and are insensitive to environmental changes are ideal for recognition. The search for invariants (e.g., algebraic and projective invariants) is a classical problem in mathematics dating back to the 18th century. The need for invariant image descriptors has long been recognized in computer vision [3, 6]. Invaria...
متن کاملIllumination Invariants
In this paper we show how tools from group theory can be used to find invariants for time-changing illumination changes. In the first part of the paper we give a brief overview over the theory of one-parameter groups and their invariants. In the second part we describe how the chromaticity coordinates of time-changing illumination spectra can be described by one-parameter curves. Examples for b...
متن کاملGroup Theoretical Invariants in Color Image Processing
Many image formation processes are complex interactions of several sub-processes and the analysis of the resulting images requires often to separate the influence of these sub-processes. An example is the formation of a color image which depends on the illumination, the properties of the camera and the objects in the scene, the imaging geometry and many other factors. Color constancy algorithms...
متن کاملA Composite Illumination Invariant Color Feature and Its Application to Partial Image Matching
In camera-based object recognition and classification, surface color is one of the most important characteristics. However, apparent object color may differ significantly according to the illumination and surface conditions. Such a variation can be an obstacle in utilizing color features. Geusebroek et al.’s color invariants can be a powerful tool for characterizing the object color regardless ...
متن کاملA Composite Illumination Invariant Color Feature and Its Application to Partial Image Matching (Draft version)
In camera-based object recognition and classification, surface color is one of the most important characteristics. However, apparent object color may differ significantly according to the illumination and surface conditions. Such a variation can be an obstacle in utilizing color features. Geusebroek et al.’s color invariants can be a powerful tool for characterizing the object color regardless ...
متن کاملA Statistical Approach for Learning Invariants: Application to Image Color Correction and Learning Invariants to Illumination
This paper presents a new approach for automatic image color correction, based on statistical learning. The method both parameterizes color independently of illumination and corrects color for changes of illumination. The motivation for using a learning approach is to deal with changes of lighting typical of indoor environments such as home and office. The method is based on learning color inva...
متن کامل