Ohio State nav bar

Seminar Series: Lo-Bin Chang

Lo-Bin Chang Seminar Series
September 3, 2020
All Day
TBA

Title

A Bayesian framework and conditional modeling for image analysis

Speaker

Lo-Bin Chang, The Ohio State University, Department of Statistics

Abstract

A forward or generative model in Bayesian image analysis consists of a prior probability distribution and a conditional data model. The prior distribution expresses common or learned knowledge about the odds of possible scene interpretations (e.g., relationships among objects or parts of objects); the conditional data model places a probability on image-based observations given any particular interpretation. To the extent that the generative model generates features, as opposed to pixel intensities, the posterior distribution (i.e., the conditional distribution on part and object labels given the image) is based on incomplete information; feature vectors are generally insufficient to recover the original image data. In this talk, I will discuss a general approach to this challenge, based on which the pixel-level models for the appearances of parts and objects can be built. For modeling scene interpretations, I will introduce a hierarchical representation to group parts and objects, and provide a perturbation formula to enhance the context sensitivity.