Context-aware classification for incremental scene interpretation

Arne Kreutzmann; Bernd Neumann; Kasim Terzic

doi:10.1145/1722156.1722158

Context-aware classification for incremental scene interpretation

Arne Kreutzmann^*, Bernd Neumann, Kasim Terzic

^*Corresponding author for this work

School of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Appearance-based classification is a difficult task in many domains due to ambiguous evidence. Knowledge about the relationships between objects in the scene can help resolve this problem. In this paper, we present a new probabilistic classification framework based on the cooperation of decision trees and Bayesian Compositional Hierarchies, and show that introducing contextual knowledge in the form of dynamic priors significantly improves classification performance in the façade domain.

Original language	English
Title of host publication	Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th Int. Conf. on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009
DOIs	https://doi.org/10.1145/1722156.1722158
Publication status	Published - 1 Dec 2009
Event	Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th International Conference on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009 - Boston, MA, United States Duration: 2 Nov 2009 → 6 Nov 2009

Conference

Conference	Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th International Conference on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009
Country/Territory	United States
City	Boston, MA
Period	2/11/09 → 6/11/09

Keywords

Context-driven event interpretation
Guided vision based on high-level reasoning

Access to Document

10.1145/1722156.1722158

Cite this

@inproceedings{b0d80b4cdc5a4b2f870f2abf492de426,

title = "Context-aware classification for incremental scene interpretation",

abstract = "Appearance-based classification is a difficult task in many domains due to ambiguous evidence. Knowledge about the relationships between objects in the scene can help resolve this problem. In this paper, we present a new probabilistic classification framework based on the cooperation of decision trees and Bayesian Compositional Hierarchies, and show that introducing contextual knowledge in the form of dynamic priors significantly improves classification performance in the fa{\c c}ade domain.",

keywords = "Context-driven event interpretation, Guided vision based on high-level reasoning",

author = "Arne Kreutzmann and Bernd Neumann and Kasim Terzic",

year = "2009",

month = dec,

day = "1",

doi = "10.1145/1722156.1722158",

language = "English",

isbn = "9781605586915",

booktitle = "Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th Int. Conf. on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009",

note = "Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th International Conference on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009 ; Conference date: 02-11-2009 Through 06-11-2009",

}

Kreutzmann, A, Neumann, B & Terzic, K 2009, Context-aware classification for incremental scene interpretation. in Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th Int. Conf. on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009., a2, Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th International Conference on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009, Boston, MA, United States, 2/11/09. https://doi.org/10.1145/1722156.1722158

Context-aware classification for incremental scene interpretation. / Kreutzmann, Arne; Neumann, Bernd; Terzic, Kasim.
Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th Int. Conf. on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009. 2009. a2.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Context-aware classification for incremental scene interpretation

AU - Kreutzmann, Arne

AU - Neumann, Bernd

AU - Terzic, Kasim

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Appearance-based classification is a difficult task in many domains due to ambiguous evidence. Knowledge about the relationships between objects in the scene can help resolve this problem. In this paper, we present a new probabilistic classification framework based on the cooperation of decision trees and Bayesian Compositional Hierarchies, and show that introducing contextual knowledge in the form of dynamic priors significantly improves classification performance in the façade domain.

AB - Appearance-based classification is a difficult task in many domains due to ambiguous evidence. Knowledge about the relationships between objects in the scene can help resolve this problem. In this paper, we present a new probabilistic classification framework based on the cooperation of decision trees and Bayesian Compositional Hierarchies, and show that introducing contextual knowledge in the form of dynamic priors significantly improves classification performance in the façade domain.

KW - Context-driven event interpretation

KW - Guided vision based on high-level reasoning

UR - http://www.scopus.com/inward/record.url?scp=77951143228&partnerID=8YFLogxK

U2 - 10.1145/1722156.1722158

DO - 10.1145/1722156.1722158

M3 - Conference contribution

AN - SCOPUS:77951143228

SN - 9781605586915

BT - Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th Int. Conf. on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009

T2 - Workshop on Use of Context in Vision Processing, UCVP 2009, in conjunction with 11th International Conference on Multimodal Interfaces and Workshop on Machine Learning for Multi-modal Interaction, ICMI-MLMI 2009

Y2 - 2 November 2009 through 6 November 2009

ER -

Context-aware classification for incremental scene interpretation

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this