TU Darmstadt / ULB / TUbiblio

Grammar-based object representations in a scene parsing task

Savova, V. and Jäkel, F. and Tenenbaum, J. B.
Taatgen, N. and van Rijn, H. (eds.) :

Grammar-based object representations in a scene parsing task.
[Online-Edition: http://csjarchive.cogsci.rpi.edu/proceedings/2009/papers/150...]
In: Proceedings of the 31st Annual Meeting of the Cognitive Science Society, Austin, TX. Cognitive Science Society , Austin, TX
[Conference or Workshop Item] , (2009)

Official URL: http://csjarchive.cogsci.rpi.edu/proceedings/2009/papers/150...

Abstract

This paper addresses the nature of visual representations associated with complex structured objects, and the role of these representations in perceptual organization. We use a novel experimental paradigm to probe subjects intuitions about parsing a scene consisting of overlapping two-dimensional objects. The objects are generated from an abstract 2-dimensional image grammar, which specifies the set of possible configurations of object parts. We show that participants' performance on the task depends on prior experience with the object class, and based on structural cues. This indicates that structural representations exerted a top-down influence on parsing. To address the question of representation type, we used a computational model of object matching in conjunction with various probabilistic representational models. Our simulations indicate that grammar-based representations derived from the original grammars are superior to more restrictive exemplar-based representations in explaining human performance on this task, as well as to more inclusive, over-generalizing grammar-based representations.

Item Type: Conference or Workshop Item
Erschienen: 2009
Editors: Taatgen, N. and van Rijn, H.
Creators: Savova, V. and Jäkel, F. and Tenenbaum, J. B.
Title: Grammar-based object representations in a scene parsing task
Language: English
Abstract:

This paper addresses the nature of visual representations associated with complex structured objects, and the role of these representations in perceptual organization. We use a novel experimental paradigm to probe subjects intuitions about parsing a scene consisting of overlapping two-dimensional objects. The objects are generated from an abstract 2-dimensional image grammar, which specifies the set of possible configurations of object parts. We show that participants' performance on the task depends on prior experience with the object class, and based on structural cues. This indicates that structural representations exerted a top-down influence on parsing. To address the question of representation type, we used a computational model of object matching in conjunction with various probabilistic representational models. Our simulations indicate that grammar-based representations derived from the original grammars are superior to more restrictive exemplar-based representations in explaining human performance on this task, as well as to more inclusive, over-generalizing grammar-based representations.

Place of Publication: Austin, TX
Publisher: Cognitive Science Society
Divisions: 03 Department Human Sciences
03 Department Human Sciences > Institute for Psychology
03 Department Human Sciences > Institute for Psychology > Models of Higher Cognition
Event Title: Proceedings of the 31st Annual Meeting of the Cognitive Science Society
Event Location: Austin, TX
Date Deposited: 09 Jul 2018 09:15
Official URL: http://csjarchive.cogsci.rpi.edu/proceedings/2009/papers/150...
Export:

Optionen (nur für Redakteure)

View Item View Item