. We describe the problem of recognition under changing illumination conditions and changing viewing positions from a computational and human vision perspective. On the computational side we focus on the mathematical problems of creating an equivalence class for images of the same 3D object undergoing certain groups of transformations --- mostly those due to changing illumination, and briefly discuss those due to changing viewing positions. The computational treatment culminates in proposing a simple scheme for recognizing, via alignment, an image of a familiar object taken from a novel viewing position and a novel illumination condition. On the human vision aspect, the paper is motivated by empirical evidence inspired by Mooney images of faces that suggest a relatively high level of visual processing is involved in compensating for photometric sources of variability, and furthermore, that certain limitations on the admissible representations of image information may exist. The psycho...

We address the problem of reconstructing 3D space in a projective framework from two or more views, and the problem of artificially generating novel views of the scene from two given views (re-projection). We describe an invariance relation which provides a new description of structure, we call projective depth, which is captured by a single equation relating image point correspondences across two or more views and the homographies of two arbitrary virtual planes. The framework is based on knowledge of correspondence of features across views, is linear, extremely simple, and the computations of structure readily extends to over-determination using multiple views. Experimental results demonstrate a high degree of accuracy in both tasks - reconstruction and re-projection. Keywords---Visual Recognition, 3D Reconstruction from 2D Views, Projective Geometry, Algebraic and Geometric Invariants. I. Introduction The geometric relation between objects (or scenes) in the world and their imag...

This paper addresses the problem of recovering relative structure, in the form of an invariant, from two views of a 3D scene. The invariant structure is computed without any prior knowledge of camera geometry, or internal calibration, and with the property that perspective and orthographic projections are treated alike, namely, the system makes no assumption regarding the existence of perspective distortions in the input images. We show that

We present a model for recovering the direction of heading of an observer who is moving relative to a scene that may contain self-moving objects. The model builds upon an algorithm proposed by Rieger and Lawton (1985), which is based on earlier work by Longuet-Higgins and Prazdny (1981). The algorithm uses velocity differences computed in regions of high depth variation to estimate the location of the .focus o.f ezpansion, which indicates the observer's heading direction. We relate the behavior of the proposed model to psychophysical observations regarding the ability of human observers to judge their heading direction, and show how the model can cope with self- moving objects in the environment. We also discuss this model in the broader context of a navigational system that performs tasks requiring rapid sensing and response through the interaction of simple task-specific routines.

The accuracy of depth judgments that are based on binocular disparity or structure from motion (motion parallax and object rotation) was studied in 3 experiments. In Experiment 1, depth judgments were recorded for computer simulations of cones specified by binocular disparity, motion parallax, or stereokinesis. In Experiment 2, judgments were recorded for real cones in a structured environment, with depth information from binocular disparity, motion parallax, or object rotation about the y-axis. In both of these experiments, judgments from binocular disparity information were quite accurate, but judgments on the basis of geometrically equivalent or more robust motion information reflected poor recovery of quantitative depth information. A 3rd experiment demonstrated stereoscopic depth constancy for distances of 1 to 3 m using real objects in a well-illuminated, structured viewing environment in which monocular depth cues (e.g., shading) were minimized. It has been pointed out that the geometric information supporting the perception of depth from binocular disparity is actually less determinate than that supporting the recovery of structure from object rotation or motion parallax

tive Vision and especially Prof. Ruzena Bajcsy, Henrik Christianssen, Prof. Jim Crowley, Prof. Randal Nelson and Prof. Giulio Sandini were most useful in the development of my ideas. The help of Kourosh Pahlavan and Prof. Jan-Olof Eklundh in gathering image data with the KTH-head is highly appreciated. Especially I would like to thank my family, Willibald and Dietlinde, Barbara, Elke, Wolfgang and Magdalena for their love and support throughout the years. This work would not have been possible without the generous support of the Osterreichisches Bundesministerium fur Wissenschaft und Forschung, the Osterreichische Bundekammer der Gewerblichen Wirtschaft and the Directorate of Robotics and Machine Intelligence of the National Science Foundation. i Contents 1 Introduction 1 1.1 Classical computer vision : : : : : : : : : : : : : : : : : : : : : : : : : : : 2 1.2 The state of the art : : : : : : : : : : : : : : : : : : : : : : : : : : :