TU Darmstadt / ULB / TUbiblio

Topological Sparse Learning of Dynamic Form Patterns

Guthier, Thomas and Willert, Volker and Eggert, Julian (2015):
Topological Sparse Learning of Dynamic Form Patterns.
In: Neural Computation, MIT Press, pp. 42-73, 27, (1), [Article]

Abstract

Motion is a crucial source of information for a variety of tasks in social interactions. The process of how humans recognize complex articulated movements such as gestures or face expressions remains largely unclear. There is an ongoing discussion if and how explicit low-level motion information, such as optical flow, is involved in the recognition process. Motivated by this discussion, we introduce a computational model that classifies the spatial configuration of gradient and optical flow patterns. The patterns are learned with an unsupervised learning algorithm based on translation-invariant nonnegative sparse coding called VNMF that extracts prototypical optical flow patterns shaped, for example, as moving heads or limb parts. A key element of the proposed system is a lateral inhibition term that suppresses activations of competing patterns in the learning process, leading to a low number of dominant and topological sparse activations. We analyze the classification performance of the gradient and optical flow patterns on three real-world human action recognition and one face expression recognition data set. The results indicate that the recognition of human actions can be achieved by gradient patterns alone, but adding optical flow patterns increases the classification performance. The combined patterns outperform other biological-inspired models and are competitive with current computer vision approaches.

Item Type: Article
Erschienen: 2015
Creators: Guthier, Thomas and Willert, Volker and Eggert, Julian
Title: Topological Sparse Learning of Dynamic Form Patterns
Language: English
Abstract:

Motion is a crucial source of information for a variety of tasks in social interactions. The process of how humans recognize complex articulated movements such as gestures or face expressions remains largely unclear. There is an ongoing discussion if and how explicit low-level motion information, such as optical flow, is involved in the recognition process. Motivated by this discussion, we introduce a computational model that classifies the spatial configuration of gradient and optical flow patterns. The patterns are learned with an unsupervised learning algorithm based on translation-invariant nonnegative sparse coding called VNMF that extracts prototypical optical flow patterns shaped, for example, as moving heads or limb parts. A key element of the proposed system is a lateral inhibition term that suppresses activations of competing patterns in the learning process, leading to a low number of dominant and topological sparse activations. We analyze the classification performance of the gradient and optical flow patterns on three real-world human action recognition and one face expression recognition data set. The results indicate that the recognition of human actions can be achieved by gradient patterns alone, but adding optical flow patterns increases the classification performance. The combined patterns outperform other biological-inspired models and are competitive with current computer vision approaches.

Journal or Publication Title: Neural Computation
Volume: 27
Number: 1
Publisher: MIT Press
Divisions: 18 Department of Electrical Engineering and Information Technology > Institut für Automatisierungstechnik und Mechatronik > Control Methods and Robotics
18 Department of Electrical Engineering and Information Technology > Institut für Automatisierungstechnik und Mechatronik
18 Department of Electrical Engineering and Information Technology
Date Deposited: 09 Oct 2014 09:18
Related URLs:
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item