Thesis advisor: Sean P. MacEvoy / For humans, healthy and productive living depends on navigating through the world and behaving appropriately along the way. But in order to do this, humans must first recognize their visual surroundings. The technical difficulty of this task is hard to comprehend: the number of possible scenes that can fall on the retina approaches infinity, and yet humans often effortlessly and rapidly recognize their surroundings. Understanding how humans accomplish this task has long been a goal of psychology and neuroscience, and more recently, has proven useful in inspiring and constraining the development of new algorithms for artificial intelligence (AI). In this thesis I begin by reviewing the current state of scene recognition research, drawing upon evidence from each of these areas, and discussing an unchallenged assumption in the literature: that scene recognition emerges from independently processing information about scenes’ local visual features (i.e. the kinds of objects they contain) and global visual features (i.e., spatial parameters. ). Over the course of several projects, I challenge this assumption with a new framework for scene recognition that indicates a crucial role for information sharing between these resources. Development and validation of this framework will expand our understanding of scene recognition in humans and provide new avenues for research by expanding these concepts to other domains spanning psychology, neuroscience, and AI. / Thesis (PhD) — Boston College, 2016. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Psychology.

Cette thèse traite du problème de localisation et contrôle de robots humanoïdes par rapport à leur environnement, tel qu'observé par ses capteurs embarqués. Le SLAM visuel dense, consistant en l'estimation simultanée d'une carte 3D de l'environnement et de la position du robot dans cette carte est exploité pour étendre et robustifier les méthodes de planification contrôle multi-contact. Celles-ci consistent à établir et exploiter des contacts robot-environnement pour accomplir des tâches de locomotion et manipulation. Des incertitudes sur la posture initiale du robot, ainsi que des perturbations causées par une modélisation inadéquate des contacts, ainsi que des perturbations externes oblige à la prise en compte de l'état du robot et son environnement. Une méthode de calibration corps-complet est également proposée, afin d'obtenir une connaissance fiable de la chaîne cinématique du robot, nécessaire pour réaliser de telles tâches. Finalement, une méthode de marche basée sur de la commande prédictive de modèles est robustifiée par la prise en compte de large perturbations, permettant d'ajuster les trajectoires de pied et du centre de masse afin de garantir sa stabilité, tout en accomplissant les objectifs désirés. Les méthodes proposées sont illustrées et validées par de multiples expérimentations sur les robots humanoïdes HRP-2Kai et HRP-4. / This thesis deals with the problem of localizing and controlling humanoid robots with respect to its environment, as observed by its on-board sensors. Dense visual SLAM, consisting in the simultaneous estimation of a 3D map of the environment and of the robot localization within that maps is exploited to extend and robustify multi-contact planning and control. Establishing and exploiting robot-environment contacts allows the accomplishment of both locomotion and manipulation tasks. Uncertainties in the initial robot posture, and perturbations arising from improper contact-modelling and external causes are accounted for by observing the state of the robot and its environment. A whole-body calibration method is also proposed, so that robust knowledge of the robot's kinematic structure is known, a prerequisite to all robot-environment interaction tasks. Finally, a walking method based on model predictive control is robustified by taking into account large perturbations, and adjusting the footstep and center-of-mass trajectories accordingly to guarantee stability while accomplishing desired objectives.Several experiments on an HRP-2Kai and an HRP-4 humanoid robots are presented and discussed to illustrate and validate each of the proposed methods.

Cette thèse étudie l'indexation et le mécanisme d'expansion de requête en recherche d'image. L'indexation sacrifie la qualité de la recherche pour une plus grande efficacité; l'expansion de requête prend ce compromis dans l'autre sens : il améliore la qualité de la recherche avec un coût en complexité additionnel. Nous proposons des solutions pour les deux approches qui utilisent une représentation continue d'un ensemble de vecteurs. Pour l'indexation, notre solution est basée sur le test par groupe. Chaque vecteur image est assigné à un groupe, et chaque groupe est représenté par un seul vecteur. C'est la représentation continue de l'ensemble des vecteur du groupe. L'optimisation de cette représentation pour produire un bon test d'appartenance donne une solution basée sur la pseudo-inverse de Moore-Penrose. Elle montre des performances supérieures à celles d'une somme basique des vecteurs du groupe. Nous proposons aussi une alternative suivant au plus près les vecteurs-images de la base. Elle optimise conjointement l'assignation des vecteurs images à des groupes ainsi que la représentation vectorielle de ces groupes. La deuxième partie de la thèse étudie le mécanisme d'expansion de requête au moyen d'un graphe pondéré représentant les vecteurs images. Cela permet de retrouver des images similaires le long d'une même variété géométrique, mais éloignées en distance Euclidienne. Nous donnons une implémentation ultra-rapide de ce mécanisme en créant des représentations vectorielles incorporant la diffusion. Ainsi, le mécanisme d'expansion se réduit à un simple produit scalaire entre les représentations vectorielles lors de la requête. Les deux parties de la thèse fournissent une analyse théorique et un travail expérimental approfondi utilisant les protocoles et les jeux de données standards en recherche d'images. Les méthodes proposées ont des performances supérieures à l'état de l'art. / In this thesis, we study the indexing and query expansion problems in image retrieval. The former sacrifices the accuracy for efficiency, whereas the latter takes the opposite perspective and improves accuracy with additional cost. Our proposed solutions to both problems consist of utilizing continuous representations of a set of vectors. We turn our attention to indexing first, and follow the group testing scheme. We assign each dataset vector to a group, and represent each group with a single vector representation. We propose memory vectors, whose solution is optimized under the membership test hypothesis. The optimal solution for this problem is based on Moore-Penrose pseudo-inverse, and shows superior performance compared to basic sum pooling. We also provide a data-driven approach optimizing the assignment and representation jointly. The second half of the transcript focuses on the query expansion problem, representing a set of vectors with weighted graphs. This allows us to retrieve objects that lie on the same manifold, but further away in Euclidean space. We improve the efficiency of our technique even further, creating high-dimensional diffusion embeddings offline, so that they can be compared with a simple dot product in the query time. For both problems, we provide thorough experiments and analysis in well-known image retrieval benchmarks and show the improvements achieved by proposed methods.

<p>The thesis covered the creation and testing of a low cost and modular sorting system of pegs used in products by Lafayette Instruments. The system is designed to check peg dimensions through use of computer vision while sorting out nonconforming parts and counting ones that are conforming. Conforming parts are separated into bins of predetermined quantities so that they do not need manual counting. The developed system will save engineers and technicians at Lafayette instruments many man hours from manually sorting and counting the roughly 160,000 pegs a year. The system will be able to sort and count at a speed comparable to a human operator while achieving an overall average accuracy of 95% or higher.</p>

A fundamental problem in computer vision is to partition an image into meaningful segments. While image segmentation is required by many applications, the thesis focuses on segmentation of computed tomography (CT) images for analysis and quality control of composite materials. The key research contribution of this thesis is a novel image segmentation framework for including end-users in computation. This represents a departure from the traditional methods, which segment images without considering domain knowledge, and access to user feedback. Given a set of CT images of three different composite materials, we would like to create a database of annotated images for all the regions of interest. The annotated images can be used to check the accuracy of segmentation algorithms. Because of how time consuming and mundane image annotation is for a person to do, we propose to turn this task into a game. The game is aimed at making the annotation task easier, because it engages imagination, creativity, fellowship of all subjects involved. In particular, we are interested in games that can be played on the internet by many people like those in Amazon Turk, so that the broader public can get involved. We create a Game with a Purpose (GWAP) called ESP 2.0 for creating image annotations, and thus enable benchmarking of existing segmentation algorithms on our database. / Graduation date: 2012

A standardized color code is used to aid the driver in the prompt recognition of highway signs. At night, these signs are illuminated by various light sources including the headlights and other fixed light sources. These light sources may distort the appearance of the colors of the signs at night. The first objective of this study was to provide information about human capabilities with respect to the recognition of different colors under daytime and nighttime lighting. The second objective was to examine the effect of changing the specifications for highway colors from the Federal Highway Administration (FHWA) standards to the American National Standards (ANSI) safety color specifications. A laboratory experiment was conducted in which subjects named the perceived colors of retroreflective signs viewed under daytime and nighttime lighting. Forty subjects from four different age groups representing the driving population participated in the study. Three color samples (red, orange, and yellow) in three different grades (engineering grade, high intensity grade, and diamond grade), and two different color specifications (FHWA and ANSI) were used. Four different fixed light sources (clear mercury, coated mercury, coated metal halide, and high pressure sodium) were used for illuminating the signs. For the nighttime condition, two headlights were used (metal halide and tungsten halogen). Daytime lighting was simulated using a fluorescent D-65 light source. Response times as well as correct responses for naming the colors were collected for each subject. Significant differences were found for nighttime versus daytime viewing of the signing materials. In general, for nighttime viewing, red and orange colors were identified faster than yellow color samples. In most cases, FHWA colors were identified significantly more accurately and faster than the ANSI colors. The coated metal halide headlight performed better than the other fixed light sources. There was no significant difference found between the tungsten halogen and the metal halide headlights. / Graduation date: 1991

Purpose: An important challenge for eye care practitioners is meeting the needs of an ever-increasing elderly population. Standard vision tests are inadequate for determining performance in real life situations. One test that was developed to address this issue is the Attended Field of View (AFOV) test (Coeckelbergh et al, 2004). This test was designed to assess the functional field of view when people are allowed to make habitual head and eye movements. The original AFOV test is no longer available. This research seeks to develop a replacement AFOV test and to demonstrate its reliability as an assessment tool. Methods: Two groups of participants were recruited. The first group consisted of seven participants between the ages of 15-41 years. The second group consisted of seven participants between the ages of 59–79 years. All subjects had visual acuities equal or better than 20/25 and no history of visual field loss. A computer-generated display was observed from a 60cm distance. The display consisted of 24 white circles on a gray background and one open circle (target). The circles were organized with one circle in the centre and eight located radially at three eccentricities (4, 8, and 12 degrees). Participants were required to locate the target circle and identify the gap direction. A response was considered correct when both the location and gap direction were accurate. Using a weighted staircase method based on presentation time each location was evaluated independently. Viewing efficiency [log (1/threshold presentation time)] was obtained for each location. The data was analyzed using repeated measures ANOVA. Results: A comparison of viewing efficiency for the two age groups demonstrates that viewing efficiency is consistently lower for the older group at all three visits. The main effect of age was observed (F1,12=25.842;p=0.000). In the older group, a significant difference was found between the second and third visits. This difference was not found in the younger group. A main effect of eccentricity was found in both groups (F2,36=30.84;p<0.000), but no interaction was observed between eccentricity and group (F2,36=0.42;p=0.662). Viewing efficiency values in the older group were lower in all directions (main effect of age) (F1,96=150.36;p<0.000). Directional variations in viewing efficiency were observed showing higher values in the horizontal axes (directions Right and Left) than along the vertical axes (directions Up and Down) in both groups. A comparison of superior and inferior hemifield data shows consistent differences for both age groups. The superior hemifield (average of directions located superiorly to the horizontal axis) demonstrate higher viewing efficiency values (better performance) than the inferior hemifield. Conclusions: The use of the new AFOV test requires a practice time before its use in order to avoid the confound of a learning effect, but subsequent data is reliable in young people. The learning effect was more significant in older people and for this reason the use of the test should be preceded by a longer practice session in this population. When interpreting the results of this test one must account for eccentricity, direction, and age.

Vision provides relevant information for safe locomotion in a variety of environments. During stair locomotion visual information may be important to detect step boundaries, transitions between ground level and stairs, handrail location, and potential hazards. Although there is a large body of literature on the role of vision during locomotion there is relatively little focused on how visual information is used during stair walking. Stairs are related to a significant number of accidents in daily living, and many of these accidents are attributed to visual factors. Therefore, understanding the role of vision during stair walking could provide insight into the mechanisms involved in stair accidents. The purpose of this thesis was to investigate the properties of the visual input used to guide locomotion on stairs. Study 1 was design to describe the gaze patterns during stair locomotion with a specific focus on transitions and handrails. Study 2 investigated the effects of performing concurrent visual and non-visual tasks on walking performance and associated gaze behaviour during stair ascent. Study 3 explored the role of peripheral visual information during visual and non-visual dual tasking. Finally, Study 4 investigated the effects of restricting the lower peripheral visual field to walk on stairs. Studies relied on the measurement in health young adults of: gaze behaviour using an eye tracker, temporal characteristics of walking using foot switches, and reaction time and errors of dual task performance. Overall, the findings of these studies highlight the importance of the lower visual field in guiding stair locomotion and the specific importance for stair transitions. Moreover, foveal vision is not specifically critical to detecting handrails or steps. Results are interpreted in the light of the specialization of the dorsal ventral stream in processing peripheral visual field information. Findings of this thesis provide basic understanding on the role of vision for stair navigation with potential applications in stair-related accident prevention programs and stair design.

Thesis (Ph. D.)--University of California, San Diego, and San Diego State University, 2004. / Vita. Includes bibliographical references.

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references (leaves 64-66). Also available in electronic version. Access restricted to campus users.