Diss. (sammanfattning) Stockholm : Karol. inst., 2003. / Härtill 5 uppsatser.

Thesis (Ph.D.)--University of Alabama at Birmingham, 2010. / Title from PDF title page (viewed on July 8, 2010). Includes bibliographical references (p.169-183).

Thesis (Ph. D.)--University of California, Santa Cruz, 1989. / Typescript. Includes bibliographical references.

The eyes are one of the most expressive non-verbal tools a person has and they are able to communicate a great deal to the outside world about the intentions of that person. Being able to decipher these communications through robust and non-intrusive gaze tracking techniques is increasingly important as we look toward improving Human-Computer Interaction (HCI). Traditionally, devices which are able to determine a user's gaze are large, expensive and often restrictive. This work investigates the prospect of using common mobile devices such as tablets and phones as an alternative means for obtaining a user's gaze. Mobile devices now often contain high resolution cameras, and their ever increasing computational power allows increasingly complex algorithms to be performed in real time. A mobile solution allows us to turn that device into a dedicated portable gaze-tracking device for use in a wide variety of situations. This work specifically looks at where the challenges lie in transitioning current state-of-the-art gaze methodologies to mobile devices and suggests novel solutions to counteract the specific challenges of the medium. In particular, when the mobile device is held in the hands fast changes in position and orientation of the user can occur. In addition, since these devices lack the technologies typically ubiquitous to gaze estimation such as infra-red lighting, novel alternatives are required that work under common everyday conditions. A person's gaze can be determined through both their head pose as well as the orientation of the eye relative to the head. To meet the challenges outlined a geometric approach is taken where a new model for each is introduced that by design are completely synchronised through a common origin. First, a novel 3D head-pose estimation model called the 2.5D Constrained Local Model (2.5D CLM) is introduced that directly and reliably obtains the head-pose from a monocular camera. Then, a new model for gaze-estimation is introduced -- the Constrained Geometric Binocular Model (CGBM), where the visual ray representing the gaze from each eye is jointly optimised to intersect a known monitor plane in 3D space. The potential for both is that the burden of calibration is placed on the camera and monitor setup, which on mobile devices are fixed and can be determined during factory construction. In turn, the user requires either no calibration or optionally a one-time estimation of the visual offset angle. This work details the new models and specifically investigates their applicability and suitability in terms of their potential to be used on mobile platforms.

La vision de la profondeur ou communément appelée « vision 3D » permet d'interpréter les relations spatiales tridimensionnelles entre les objets de la scène visuelle et confère à l'homme une grande précision dans ses interactions avec l'environnement. La vision 3D repose sur de nombreux mécanismes d'analyse du signal visuel dont la plupart gardent tout leur pouvoir informationnel lors de la stimulation d'un seul œil (indices monoculaires) mais dont certains nécessitent la stimulation des deux yeux (indices binoculaires). Cette thèse se concentre sur les mécanismes nécessitant les deux yeux qui mettent en jeu la différence de point de vue entre les deux yeux, aussi appelée disparité rétinienne. Les travaux présentés dans cette thèse s'inscrivent suivant deux approches majeures : une approche par la modélisation avec la simulation de l'extraction de la disparité rétinienne au niveau cortical et une approche expérimentale avec l'étude de l'influence de la disparité rétinienne sur l'attention visuelle pendant l'exploration de scènes visuelles stéréoscopiques avec enregistrements oculométriques. Le modèle proposé est construit en utilisant les données physiologiques du cortex visuel primaire V1 disponibles dans la littérature. Il effectue une estimation de la disparité rétinienne à partir des réponses modélisées des cellules simples et complexes de V1. Nous adoptons une approche bio-inspirée à deux niveaux. Le premier concerne l'architecture globale d'organisation et d'interaction des cellules corticales réalisant l'extraction d'informations à différentes fréquences spatiales, orientations et disparités, pour obtenir une estimation locale de la disparité rétinienne. Le second niveau concerne la configuration des cellules corticales implémentées comme des opérateurs de filtrage spatial. La partie expérimentale se divise elle-même en deux parties. En effet, l'utilisation de données oculométriques d'exploration de scènes 3D nécessite une étape préalable de calibration. Ainsi, nous développons une méthode de calibration 3D permettant de suivre la profondeur du regard à partir des coordonnées binoculaires enregistrées par oculométrie. Ensuite, nous analysons l'influence de la disparité rétinienne sur l'exploration visuelle de différentes catégories de scènes naturelles basées sur la présence d'indices de profondeur monoculaires et binoculaires. Nous étudions l'influence de la dominance oculaire, du biais de centralité et du biais de profondeur sur l'exploration pour chaque catégorie d'image. Un modèle de saillance 2D se montre inadapté pour prédire les zones saillantes en 3D mais également en 2D. L'information de profondeur doit être intégrée dans le calcul de la saillance grâce à la disparité rétinienne et grâce aux autres indices de profondeur pour expliquer pleinement l'exploration 2D et 3D. / Depth vision or « 3D vision » can interpret tridimensional spatial relations between objects in a visual scene and gives humans a good precision of interaction with their environment. 3D vision uses several kinds of mechanisms to analyze visual signal. Some keep their power during a monocular stimulation (monocular depth cues) but others need a binocular stimulation (binocular depth cues). This thesis focuses on the binocular mechanism which uses the difference of point of view between the two eyes (also called retinal disparity). The work presented in this thesis follows two main approaches: the modeling of the retinal disparity extraction at the cortical level, and an experiment to analyze the influence of retinal disparity on visual attention during the exploration of natural stereoscopic scenes with eye tracking recording. The proposed model is built from physiologic data of primary visual cortex V1 found in the literature. Our model makes an estimation of the retinal disparity from modeled responses of simple and complex cells of V1. We take a bio-inspired approach at two levels. The first level concerns the global architecture of the organization and the interaction of cortical cells which extract the information at different spatial frequencies, orientations and disparities. The second level concerns the configuration of cortical cells implemented like spatial filters. The experimental part is subdivided into two parts. Indeed, the use of eye-tracking data of 3D scenes exploration needs a calibration step. Hence, we developed a 3D calibration method allowing us to track the depth of the gaze from the recorded binocular coordinates. Then, we analyze the influence of retinal disparity in the visual exploration of different categories of natural scenes based on the presence of monocular and binocular depth cues. The ocular dominance, the central bias and the depth bias are also studied in this paradigm. We show that a 2D saliency model is not adapted to predict the salient zone during 3D viewing but also during 2D viewing. The depth information must be integrated in saliency computation thanks to retinal disparity and monocular depth cues to explain fully the visual exploration both in 2D and 3D.

Orientador: Clesio Luis Tozzi / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-03T22:31:54Z (GMT). No. of bitstreams: 1 Forster_CarlosHenriqueQuartucci_D.pdf: 1511668 bytes, checksum: 09ed5f90b9443a930cf04e79b14d1e18 (MD5) Previous issue date: 2004 / Doutorado

The Pedestrian Detection System (PDS) has become a significant area of research designed to protect pedestrians. Despite the huge number of research work, the most current PDSs are designed to detect pedestrians without knowing their distances from cars. In fact, a priori knowledge of the distance between a car and pedestrian allows this system to make the appropriate decision in order to avoid collisions. Typical methods of distance measurement require additional equipment (e.g., Radars) which, unfortunately, cannot identify objects. Moreover, traditional stereo-vision methods have poor precision in long-range conditions. In this thesis, we use the keypoint-based feature extraction method to generate the parallax in a binocular vision system in order to measure a detectable object; this is used instead of a disparity map. Our method enhances the tolerance to instability of a moving vehicle; and, it also enables binocular measurement systems to be equipped with a zoom lens and to have greater distance between cameras. In addition, we designed a crossover re-detection and tracking method in order to reinforce the robustness of the system (one camera helps the other reduce detection errors). Our system is able to measure the distance between cars and pedestrians; and, it can also be used efficiently to measure the distance between cars and other objects such as Traffic signs or animals. Through a real word experiment, the system shows a 7.5% margin of error in outdoor and long-range conditions.

<p> The study concerns how human observers judge the relative position of successively presented points of light in an otherwise dark field. In particular, the possible role of involuntary head movements and binocular/monocular viewing conditions is considered. The data are analysed in terms of a mathematical model of the perceptual process which deals with short term memory for visual position. Contrary to previous suggestions in the literature, neither of the viewing variables proved to have a significant effect. In addition, the results provide a strong test of the theoretical model which appears to confirm the model's validity. The results of this study are shown to suggest a particular direction for future experimentation.</p> / Thesis / Master of Arts (MA)

In the past decade, considerable attention has been paid to examination of the impact of amblyopia, and strabismus, upon the lives of the individuals. Although an extensive amount of literature exists regarding amblyopia and its associated visual defects, little is known about the contribution of the amblyopic eye in the habitual viewing condition (i.e. both eyes viewing). The purpose of these studies was to determine whether amblyopes are disadvantaged in the performance of tasks under habitual viewing conditions, highlighting any functional differences which may exist as a consequence of amblyopia. Secondly, the work aimed to investigate whether the amblyopic eye contributes to the habitual performance of these tasks. A simple light detection task, in a dichoptic arrangement based upon blue/yellow stimuli viewed through yellow filters, was used to investigate the above two aims and investigate the degree of interocular suppression in amblyopic participants. Using a 3D motion analysis system performance was assessed for an obstacle crossing task (adaptive gait) and a task of reaching for and grasping of an isolated object and in a ¿cluttered¿ environment. Fine motor skills were assessed in a threading a needle task. On the whole it was found that amblyopes are not disadvantaged under habitual viewing conditions, and in cases where differences were found to exist this appeared to be in tasks requiring speed and accuracy. Consistently across all studies it was found that the amblyopic eye contributed in a positive manner, thus, as in visual normals, two eyes are better than one.

No description available.