No description available.

No description available.

This thesis investigates the problem of visual learning using a robotic platform. Given a set of objects the robots task is to autonomously manipulate, observe, and learn. This allows the robot to recognise objects in a novel scene and pose, or separate them into distinct visual categories. The main focus of the work is in autonomously acquiring object models using robotic manipulation. Autonomous learning is important for robotic systems. In the context of vision, it allows a robot to adapt to new and uncertain environments, updating its internal model of the world. It also reduces the amount of human supervision needed for building visual models. This leads to machines which can operate in environments with rich and complicated visual information, such as the home or industrial workspace; also, in environments which are potentially hazardous for humans. The hypothesis claims that inducing robot motion on objects aids the learning process. It is shown that extra information from the robot sensors provides enough information to localise an object and distinguish it from the background. Also, that decisive planning allows the object to be separated and observed from a variety of dierent poses, giving a good foundation to build a robust classication model. Contributions include a new segmentation algorithm, a new classication model for object learning, and a method for allowing a robot to supervise its own learning in cluttered and dynamic environments.

Deep learning and convolutional neural networks have become the dominant tool for computer vision. These techniques excel at learning complicated representations from data using supervised learning. In particular, image recognition models now out-perform human baselines under constrained settings. However, the science of computer vision aims to build machines which can see. This requires models which can extract richer information than recognition, from images and video. In general, applying these deep learning models from recognition to other problems in computer vision is significantly more challenging. This thesis presents end-to-end deep learning architectures for a number of core computer vision problems; scene understanding, camera pose estimation, stereo vision and video semantic segmentation. Our models outperform traditional approaches and advance state-of-the-art on a number of challenging computer vision benchmarks. However, these end-to-end models are often not interpretable and require enormous quantities of training data. To address this, we make two observations: (i) we do not need to learn everything from scratch, we know a lot about the physical world, and (ii) we cannot know everything from data, our models should be aware of what they do not know. This thesis explores these ideas using concepts from geometry and uncertainty. Specifically, we show how to improve end-to-end deep learning models by leveraging the underlying geometry of the problem. We explicitly model concepts such as epipolar geometry to learn with unsupervised learning, which improves performance. Secondly, we introduce ideas from probabilistic modelling and Bayesian deep learning to understand uncertainty in computer vision models. We show how to quantify different types of uncertainty, improving safety for real world applications.

A key aspect of how we understand the world revolves around an ability to manipulate our surroundings to experiment. From the scientific method through theories of child development, the ability to experiment is deemed critical; however, few studies have been performed to understand the strengths and weaknesses of different experimental strategies. This dissertation investigated the effectiveness of several different experimental strategies when learning about an unknown system. An empirical study was performed using binary functions as hypotheses and using computer programs to model several different experimental strategies. These strategies were derived from our definition of a normative experiment selector, which described how an idealized experimenter should select experiments. A detailed program of study was performed on these computer programs to determine the strengths and weaknesses of the experimental strategies they implemented. The number of experiments needed to determine a target theory from an initial set of hypotheses was measured. Two key discoveries were made. First, we discovered that simple experimental strategies were the most effective. For example, the most effective strategy we discovered was experimental relevance selecting any experiment guaranteeing elimination of at least a single hypothesis from the set being considered. Complex strategies to determine the optimal experiment to perform were very costly for a slight performance gain. Second, we discovered that only two factors had any major effect on performance: the number of experimental outcomes and the number of initial hypotheses considered. The number of experiments available to the experiment selector had little or no effect. Our best situations were where: (a) only a small number of hypotheses were possible, (b) each experiment had a large number of outcomes, and (c) relevant experiments were easy to determine and perform. / Graduation date: 1991

No description available.

Thesis (Ph.D.)--Australian National University, 2005.

Thesis (Ph. D.)--University of Texas at Austin, 2008. / Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Oregon, 2008. / Typescript. Includes vita and abstract. Includes bibliographical references (leaves 163-167). Also available online in Scholars' Bank; and in ProQuest, free to University of Oregon users.

Thesis (M.S.)--Brigham Young University. Dept. of Computer Science, 2007. / Includes bibliographical references (p. 55-57).