The Theory and Practice of Bayesian Image Labeling
Abstract
Integrating disparate sources of information has been recognized as one of the keys to the success of general purpose vision systems. Image clues of the keys to the success of general purpose vision systems. Image clues such as shading, texture, stereo disparities and image flows provide uncertain, local and incomplete information about the three-dimensional scene. Spatial a priori knowledge plays the role of filling in missing information and smoothing out noise. This thesis proposes a solution to the longstanding open problem of visual integration. It reports a framework, based on Bayesian probability theory, for computing an intermediate representation of the scene from disparate sources of information. The computation is formulated as labeling problem. Local visual observations for each image entity are reported as label likelihoods. They are combined consistently and coherently an hierarchically structured label trees with a new, computationally simple procedure. The pooled label likelihoods are fused with the a priori spatial knowledge encoded as Markov Random Fields (MRF's) The posteriori distribution of the labelings are thus derived in a Bayesian formalism. A new inference method, Highest Confidence First (HCF) estimation, is used to infer a unique labeling from the posteriori distribution.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1988
- Accession Number
- ADA202037
Entities
People
- Paul B. Chou
Organizations
- University of Rochester