Tod S. Levitt

UAI Conference 1999 Conference Paper

Learning Bayesian Networks from Incomplete Data with Stochastic Search Algorithms

• James W. Myers
• Kathryn B. Laskey
• Tod S. Levitt

This paper describes stochastic search approaches, including a new stochastic algorithm and an adaptive mutation operator, for learning Bayesian networks from incomplete data. This problem is characterized by a huge solution space with a highly multimodal landscape. State-of-the-art approaches all involve using deterministic approaches such as the expectation-maximization algorithm. These approaches are guaranteed to find local maxima, but do not explore the landscape for other modes. Our approach evolves structure and the missing data. We compare our stochastic algorithms and show they all produce accurate results.

UAI Conference 1994 Conference Paper

Incremental Dynamic Construction of Layered Polytree Networks

• Keung-Chi Ng
• Tod S. Levitt

AIIM Journal 1993 Journal Article

Bayesian inference for model-based segmentation of computed radiographs of the hand

• Tod S. Levitt
• Marcus W. Hedgcock
• John W. Dye
• Scott E. Johnston
• Vera M. Shadle
• Dmitry Vosky

We present a method for medical image understanding by computer that uses model-based, hierarchical Bayesian inference to accurately segment imaged anatomy. A first application is a prototype system that automatically segments and measures symptoms of arthridities in hand radiographs. This is potentially useful in radiological diagnosis and tracking of arthridities. Key steps of the model-based, Bayesian inference approach are: (1) prediction of imagery features from 3D models of anatomy, parameterized by population statistics, (2) local image feature extraction in predicted sub-regions, and (3) the use of a probabilistic calculus to accrue results of image processing and image feature matching procedures in support or denial of hypotheses about the imaged anatomy. The prototype system for hand radiograph analysis accurately segments normal and somewhat degenerated hand anatomy. Results are shown of the ability of the automated system to ‘fail soft’, recognizing when segmentation is inadequate for accurate measurement. This self evaluation capability improves reliability of measurements for potential clinical use.

AIJ Journal 1990 Journal Article

Qualitative navigation for mobile robots

• Tod S. Levitt
• Daryl T. Lawton

UAI Conference 1989 Conference Paper

Model-Based Influence Diagrams for Machine Vision

• Tod S. Levitt
• John Mark Agosta
• Thomas O. Binford

We show an approach to automated control of machine vision systems based on incremental creation and evaluation of a particular family of influence diagrams that represent hypotheses of imagery interpretation and possible subsequent processing decisions. In our approach, model-based machine vision techniques are integrated with hierarchical Bayesian inference to provide a framework for representing and matching instances of objects and relationships in imagery and for accruing probabilities to rank order conflicting scene interpretations. We extend a result of Tatman and Shachter to show that the sequence of processing decisions derived from evaluating the diagrams at each stage is the same as the sequence that would have been derived by evaluating the final influence diagram that contains all random variables created during the run of the vision system.

UAI Conference 1988 Conference Paper

Utility-based control for computer vision

• Tod S. Levitt
• Thomas O. Binford
• Gil J. Ettinger

UAI Conference 1987 Conference Paper

Bayesian Inference in Model-Based Machine Vision

• Thomas O. Binford
• Tod S. Levitt
• Wallace B. Mann

AAAI Conference 1987 Conference Paper

Qualitative Landmark-Based Path Planning and Following

• Tod S. Levitt
• David M. Chelberg

This paper develops a theory for path planning and following using visual landmark recognition for the representation of environmental locations. It encodes local perceptual knowledge in structures called viewframes and orientation regions. Rigorous representations of places as visual events are developed in a uniform framework that smoothly integrates a qualitative version of path planning with inference over traditional metric representations. Paths in the world are represented as sequences of sets of landmarks, viewframes, orientation boundary crossings, and other distinctive visual events. Approximate headings are computed between viewframes that have lines of sight to common landmarks. Orientation regions are range-free, topological descriptions of place that are rigorously abstracted from viewframes. They yield a coordinate-free model of visual landmark memory that can also be used for path planning and following. With this approach, a robot can opportunistically observe and execute visually cued "shortcuts".

UAI Conference 1986 Conference Paper

Bayesian inference for radar imagery based surveillance

• Tod S. Levitt

ICRA Conference 1986 Conference Paper

Terrain models for an autonomous land vehicle

• Daryl T. Lawton
• Tod S. Levitt
• Christopher C. McConnell
• Jay Glicksman

We present an architecture for terrain recognition for an autonomous land vehicle. Basic components of this are a set of data bases for generic object models, perceptual structures, temporary memory for the instantiation of object and relational hypothesis, and a long term memory for storing stable hypothesis which are affixed to the terrain representation. Different inference processes operate over these data bases. We describe components of this architecture: the perceptual structure data base, the grouping processes that operate over this, and schemas. We conclude with a processing example for matching predictions from the long term terrain model to imagery and extracting significant perceptual structures for consideration as potential landmarks.

UAI Conference 1985 Conference Paper

Model-Based Probabilistic Situation Inference in Hierarchical Hypothesis Spaces

• Tod S. Levitt