Edinburgh Research Explorer

Dynamic Trees: Learning to Model Outdoor Scenes

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publicationComputer Vision — ECCV 2002
Subtitle of host publication7th European Conference on Computer Vision Copenhagen, Denmark, May 28–31, 2002 Proceedings, Part IV
PublisherSpringer Berlin Heidelberg
Pages82-96
Number of pages15
ISBN (Electronic)978-3-540-47979-6
ISBN (Print)978-3-540-43748-2
DOIs
Publication statusPublished - 2002

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Berlin Heidelberg
Volume2353
ISSN (Print)0302-9743

Abstract

This paper considers the dynamic tree (DT) model, first introduced in [1]. A dynamic tree specifies a prior over structures of trees, each of which is a forest of one or more tree-structured belief networks (TSBN). In the literature standard tree-structured belief network models have been found to produce "blocky" segmentations when naturally occurring boundaries within an image did not coincide with those of the subtrees in the fixed structure of the network. Dynamic trees have a flexible architecture which allows the structure to vary to create configurations where the subtree and image boundaries align, and experimentation with the model has shown significant improvements.

ID: 21873025