Undergraduate honors paper--Mount Holyoke College, 2006. Dept. of Computer Science. / Includes bibliographical references (leaves 80-82).

Thesis (M.S.)--Eastern Illinois University, 1987. / Includes bibliographical references (leaves 31-32).

M. Ing. / The process of extracting depth information from multiple two-dimensional images taken of the same scene is known as stereo vision. It is of central importance to the field of machine vision as it is a low level task required for many higher level applications. The past few decades has witnessed the development of hundreds of different stereo vision algorithms. This has made it difficult to classify and compare the various approaches to the problem. In this research we provide an overview of the types of approaches that exist to solve the problem of stereo vision. We focus on a specific subset of algorithms, known as local stereo algorithms. Our goal is to critically analyse and compare a representative sample of local stereo algorithm in terms of both speed and accuracy. We also divide the algorithms into discrete interchangeable components and experiment to determine the effect that each of the alternative components has on an algorithm’s speed and accuracy. We investigate even further to quantify and analyse the effect of various design choices within specific algorithm components. Finally we assemble all of the knowledge gained through the experimentation to compose and optimise a novel algorithm. The experimentation highlighted the fact that by far the most important component of a local stereo algorithm is the manner in which it aggregates matching costs. All of the top performing local stereo algorithms dynamically define the shape of the windows over which the matching costs are aggregated. This is done in a manner that aims to only include pixels in a window that is likely to be at the same depth as the depth of the centre pixel of the window. Since the depth is unknown, the cost aggregation techniques use colour and proximity information to best guess whether pixels are at the same depth when defining the shape of the aggregation windows. Local stereo algorithms are usually less accurate than global methods but they are supposed to be faster and more parallelisable. These cost aggregation techniques result in very accurate depth estimates but unfortunately they are also very expensive computationally. We believe the focus of local stereo algorithm development should be speed. Using the experimental results we developed an algorithm that achieves accuracies in the same order of magnitude as the state-of-the-art algorithms while reducing the computation time by over 50%.

No description available.

This research studied egocentric depth perception in an augmented reality (AR) environment. Specifically, it involved measuring depth perception in the near visual field by using quantitative methods to measure the depth relationships between real and virtual objects. This research involved two goals; first, engineering a depth perception measurement apparatus and related calibration andmeasuring techniques for collecting depth judgments, and second, testing its effectiveness by conducting an experiment. The experiment compared two complimentary depth judgment protocols: perceptual matching (a closed-loop task) and blind reaching (an open-loop task). It also studied the effect of a highly salient occluding surface; this surface appeared behind, coincident with, and in front of virtual objects. Finally, the experiment studied the relationship between dark vergence and depth perception.

No description available.

Augmented Reality (AR) is a method of mixing computer-generated graphics with real-world environments. In AR, observers retain the ability to see their physical surroundings while additional (augmented) information is depicted as simulated graphical objects matched to the real-world view. In the following experiments, optical see-through head-mounted displays (HMDs) were used to present observers with both Augmented and Virtual Reality environments. Observers were presented with varied real, virtual, and combined stimuli with and without the addition of motion parallax. The apparent locations of the stimuli were then measured using quantitative methods of egocentric depth judgment. The data collected from these experiments were then used to determine how observers perceived egocentric depth with respect to both real-world and virtual objects.

No description available.

Many systems today do not allow the operator to view the environment directly. For example, operators of telerobotic or remote manipulation systems may be far removed from the work site. In these situations, images from cameras at the task site provide the only visual link between the task and operator. Such imaging systems are unable to provide exact representations of the task environment, resulting in a possible degradation of important visual cues. If visual cues are degraded, task performance may be less accurate, require increased completion time, and subject the operator to hazardous conditions. In this study, the field of view and contrast of imagery were manipulated to determine their effects on observers' ability to estimate distances of targets within natural settings. Images were video taped with a rotating camera and the targets were placed between 9.14 m and 60.96 m from the camera. The fields of view ranged from 20 deg to 74 deg. A lightly wooded area and an open field were used to provide either high or low visual contexts, respectively. Participants rated their confidence in the distance estimates on a seven-point scale. / Master of Science

No description available.