In this thesis, we describe a statistical method for 3D object detection. In this method, we decompose the 3D geometry of each object into a small number of viewpoints. For each viewpoint, we construct a decision rule that determines if the object is present at that specific orientation. Each decision rule uses the statistics of both object appearance and "non-object " visual appearance. We represent each set of statistics using a product of histograms. Each histogram represents the joint statistics of a subset of wavelet coefficients and their position on the object. Our approach is to use many such histograms representing a wide variety of visual attributes. Using this method, we have developed the first algorithm that can reliably detect faces that vary from frontal view to full profile view and the first algorithm that can reliably detect cars over a wide range of viewpoints.

This paper presents a generic method for perceptual grouping, and an analysis of its expected grouping quality. The grouping method is fairly general: it may be used for the grouping of various types of data features, and to incorporate different grouping cues, operating over feature sets of different sizes. The proposed method is divided into two parts: Constructing a graph representation of the available perceptual grouping evidence, and then finding the "best" partition of the graph into groups. The first stage includes a cue enhancement procedure, which integrates the information available from multi-feature cues into very reliable bi-feature cues. Both stages are implemented using known statistical tools such as Wald's SPRT algorithm and the Maximum Likelihood criterion. The accompanying theoretical analysis of this grouping criterion quantifies intuitive expectations and predicts that the expected grouping quality increases with cue reliability. It also shows that investing more computational effort in the grouping algorithm leads to better grouping results. This analysis, which quantifies the grouping power of the Maximum Likelihood criterion, is independent of the grouping domain. To our best knowledge, such an analysis of a grouping process is given here for the first time. Three grouping algorithms, in three different domains, are synthesized as instances of the generic method, They demonstrate the applicability and generality of this grouping method. Keywords : Perceptual Grouping, Grouping Analysis, Graph Clustering, Maximum Likelihood, Wald's SPRT, Performance Prediction, Generic Grouping Algorithm. 1

The likelihood principle states that the visual system prefers the most likely interpretation of a stimulus, whereas the simplicity principle states that it prefers the most simple interpretation. This study investi-gates how close these seemingly very different principles are by combining findings from classical, algorithmic, and structural information theory. It is argued that, in visual perception, the two principles are perhaps very different with respect to the viewpoint-independent aspects of perception but probably very close with respect to the viewpoint-dependent aspects which, moreover, seem decisive in everyday perception. This implies that either principle may have guided the evolution of visual systems and that the simplicity paradigm may provide perception models with the necessary quantitative specifications of the often plausible but also intuitive ideas provided by the likelihood paradigm. In visual perception research, an ongoing debate concerns the question of whether the likelihood principle (Von Helmholtz, 1909/1962) or the simplicity principle (Hochberg & McAlister, 1953) provides the best explanation of the human interpretation of visual stimuli. The phenomenon to be explained is, more specifi-cally, that human subjects usually show a clear preference for only

Meaningful objects in a scene move with purpose. The ability to induce visual expectations from such purpose is important in visual observation. By regarding the spatio-temporal regularities in the moving patterns of an object in the scene as a network of temporally dependent belief hypothesis, visual expectations can be represented by the most likely combinations of the hypotheses based on updating the network in response to instantaneous visual evidence. A particular type of probabilistic single path Directed Acyclic Graph (DAG) belief network, the Hidden Markov Model (HMM), can be used to represent the "hidden" regularities behind the apparently random moves of an object in a scene and reproduce such regularities as "blind", therefore, insensitive expectations. By adaptively adjusting such a probabilistic belief network with observed visual evidence instantaneously, a Visual Augmented Hidden Markov Model (VAHMM) can be used to model and produce dynamic expectations of a moving objec...

This paper proposes a new, efficient, figure from ground discrimination method. This algorithm is based on the assumption that background data features can be more easily detected than figure data features, thus emphasizing the background detection task (and implying the name of the method). Along the iterative labeling process, data features are sequentially and permanently labelled as "background", while more global information is being collected to assist with the coming decisions, until the process converges. This procedure creates a bootstrap mechanism which improves performance in very cluttered scenes. The method can be applied to many perceptual grouping cues, and an application to smoothness-based classification of edge points is given. A fast implementation using a kd-tree allows to work on large, realistic images.

Software visualization is nifty stuff; but is it the powerful cognitive tool it is often assumed to be? This chapter attempts to moderate the understandable enthusiasm for software visualization and to raise some of the questions for which the discipline doesn't yet have answers. The chapter is structured as a list of questions with discussion. The questions are not a comprehensive analysis of cognitive challenges in software visualization. Rather, the chapter attempts to provide a list sufficiently provocative to give designers pause, in order: (a) to establish that good software visualization isn't simply a matter of mimicking paper-based tasks or doing what is technically easy---and certainly isn't `solved' yet; but also (b) even simple tools can improve software comprehension, if they're the right ones.

One of the main goals in realistic rendering is to generate images that are indistinguishable from photographs -- but how do observers decide whether an image is photographic or computer-generated? If this perceptual process were understood, then rendering algorithms could be developed to directly target these cues. In this paper we introduce an experimental method for measuring the perception of visual realism in images, and present the results of a series of controlled human subject experiments. These experiments cover the following visual factors: shadow softness, surface smoothness, number of light sources, number of objects, and variety of object shapes. This technique can be used to either affirm or cast into doubt common assumptions about realistic rendering. The experiments can be performed using either photographs or computergenerated images. This work provides a first step towards objectively understanding why some images are perceived as photographs, while others as computer graphics. 1

This paper reviews the key issues in hypermedia systems as an overture to the proposal of a new semiotic paradigm for hypermedia data and coding models. The hypertext concept permits users to interact with and manage data as high-level conceptual objects rather than as symbol streams. Current hypermedia systems can best be defined as an amalgamation of hypertext and multimedia. While the hypertext data model enables this goal, that is not true for the data models of other media forms. A new semiotic paradigm that addresses these deficiencies and supports object-oriented interaction with compressed multimedia streams is proposed. This paper initially presents an overview of the hypertext data model, contrasting it with existing multimedia data and coding models. The framework for the new paradigm is then presented in a brief review of cognitive, psychological, and semiotic principles. This analysis culminates in the proposal of semiotically based data models and representations predisposed to the hypermedia paradigm. Keywords—Audio coding, data models, hypermedia, hypertext systems, image coding, information retrieval, multimedia information systems, psychology, semiotics, signal representations, source coding. I.

The perceived realism of a computer generated image depends on the accuracy of the modeling and illumination calculations, the limitations of the display device, and the way in which the Human Visual System processes this information. A real environment is unlikely to be pristine but will have accumulated dirt, dust and scratches from everyday use. Although human observers do not perhaps consciously take note of these phenomena, the absence of such features from the synthetic representation of that real scene may indeed affect the viewer's perceived realism of the virtual environment. This paper presents a series of psychophysical experiments to examine whether perceived realism of a virtual environment may be improved by adding textures artistically enhanced. Categories and Subject Descriptors

or cells known as rods and cones due to the shapes of their photosensitive outer segments. There are somewhere between 100-120 million rods and 7-8 million cones in each retina. The rods are extremely sensitivetolight and provide achromatic vision at low(scotopic) levels of illumination. The cones are less sensitive than the rods, but provide color vision at high (photopic) levels of illumination. The photosensitive segments of the rods and cones are located closest to the choroid layer. This means that light striking the retina must first pass through several layers of neural tissue to reach the photoreceptors. Only in a small 1.5mm diameter area near the optic axis called the fovea are the cell bodies and neural fibers drawn aside so the photoreceptive surfaces are directly exposed to light. The rod and cone systems are sensitivetolightwithwavelengths from about 400nm to 700nm. The rods have their peak sensitivity