This paper provides a comprehensive survey of the technical achievements in the research area of image retrieval, especially content-based image retrieval, an area that has been so active and prosperous in the past few years. The survey includes 100+ papers covering the research aspects of image feature representation and extraction, multidimensional indexing, and system design, three of the fundamental bases of content-based image retrieval. Furthermore, based on the state-of-the-art technology available now and the demand from real-world applications, open research issues are identified and future promising research directions are suggested. C ○ 1999 Academic Press 1.

This paper provides a comprehensive survey of the technical achievements in the research area of Image Retrieval, especially Content-Based Image Retrieval, an area so active and prosperous in the past few years. The survey includes 100+ papers covering the research aspects of image feature representation and extraction, multi-dimensional indexing, and system design, three of the fundamental bases of Content-Based Image Retrieval. Furthermore, based on the state-of-the-art technology available now and the demand from real-world applications, open research issues are identified, and future promising research directions are suggested. 1. INTRODUCTION Recent years have seen a rapid increase of the size of digital image collections. Everyday, both military and civilian equipment generates giga-bytes of images. Huge amount of information is out there. However, we can not access to or make use of the information unless it is organized so as to allow efficient browsing, searching and retriev...

We describe procedures for creating efficient spectral representations for color. The representations generalize conventional tristimulus representations, which are based on the peripheral encoding by the human eye. We use low-dimensional linear models to approximate the spectral properties of surfaces and illuminants with respect to a collection of sensing devices. We choose the linear model basis functions by minimizing the error in approximating sensor responses for collections of surfaces and illuminants. These linear models offer some conceptual simplifications for applications such as printer calibration; they also perform substantially better than principal components approximations for computer graphics applications.

Most multi-camera vision applications assume a single common color response for all cameras. However different cameras—even of the same type—can exhibit radically different color responses, and the differences can cause significant errors in scene interpretation. To address this problem we have developed a robust system aimed at inter-camera color consistency. Our method consists of two phases: an iterative closed-loop calibration phase that searches for the per-camera hardware register settings that best balance linearity and dynamic range, followed by a refinement phase that computes the per-camera parametric values for an additional software-based color mapping. 1.

How can one identify specularities when an object is illuminated by light that varies in spectrum with direction from the surface? A linear model of color shading can answer this question and also recover surface orientation in non--specular regions using only a single color image of the surface taken under a set of illuminants whose positions, strengths, and spectral content need not be known a priori. The shape--from--color method is based on a Lambertian model. For such a reflectance model the surface normal is related in a linear way to the measured RGB color. Linearity means that the Gaussian sphere is transformed into an ellipsoid in color space, and one can solve for the ellipsoid using least squares; surface normals are recovered only up to an overall orthogonal transformation unless additional constraints are employed. When specularities are present, least squares no longer works. If, however, one views specularities as outliers to the underlying color ellipsoid, then a robust...

this paper.

Abstract. We propose a new method for the Lambertian Shape From Shading (SFS) problem based on the notion of Crandall-Lions viscosity solution. This method has the advantage of requiring the knowledge of the solution (the surface to be reconstructed) only on some part of the boundary and/or of the singular set (the set of the points at maximal intensity). Moreover it unifies in an unique mathematical formulation the works of Rouy et al. [50, 34], Falcone et al. [21], Prados et al. [49, 46, 48], based on the notion of viscosity solutions and the work of Dupuis and Oliensis [17] dealing with classical solutions and value functions. Also, it allows to generalize their results to the “perspective SFS ” problem recently simultaneously introduced in [46, 55, 13]. While the theoretical part has been developed in [44], in this paper we give some stability results and we describe numerical schemes for the SFS based on this method. We construct provably convergent and robust algorithms. Finally, we apply our SFS method to real images and we suggest some real-life applications.

Statistics-based colour constancy algorithms work well as long as there are many colours in a scene, they fail however when the en- countering scenes comprise few surfaces. In contrast, physics-based algorithms, based on an understanding of physical processes such as highlights and interreflections, are theoretically able to solve for colour constancy even when there are as few as two surfaces in a scene. Un- fortunately, physics-based theories rarely work outside the lab. In this paper we show that a combination of physical and statistical knowledge leads to a surprisingly simple and powerful colour constancy algorithm, one that also works well for images of natural scenes.

It is often possible and desirable to reduce the dimensionality of variables describing data, e.g. when the original number of variables is large and when errors introduced by approximation can be tolerated. This is particularly the case with respect to spectral measurements of illuminants and surfaces. In computer graphics and computer vision, it is often the case that multiplications of whole spectra, component by component, must be carried out. For example, this is the case when light reflects from matter, or is transmitted through materials. This statement particularly holds for spectrally-based ray tracing or radiosity in graphics. There, many such multiplications must be carried out, making a full-spectrum method prohibitively expensive. However, using full spectra is attractive because of the many important phenomena that can only be modelled using all the physics at hand. Here we apply to the task of spectral multiplication a method previously used in modeling RGB-based light propagation. We will in fact show that we can often multiply spectra without carrying out spectral multiplication.

When a Lambertian surface is illuminated by several chromatic lights the surface normals may berecovered from a single color image. A robust regression is used to #nd the ellipsoid in color space on which at least half the pixels lie. Then the matrix giving the linear relationship between the color and the surface normal, for non-outlier points, is found as a root of the ellipsoid quadratic form. But this root is recovered only up to an arbitrary rotation. An integrability condition can be used to determine the correct rotation. The rotation of recovered surface normals is needed to align partial derivatives p and q with the camera plane and thus establish the object's attitude. Here a new smoothness condition approximating the integrability condition is introduced that allows one to solve for the rotation matrix in closed form.