Event models for recognition and natural language description of events in real-world image sequences

Authors:
Bernd Neumann;Hans-Joachim Novak
Affiliations:
Fachbereich Informatik, Hamburg, Germany;Fachbereich Informatik, Hamburg, Germany
Venue:
IJCAI'83 Proceedings of the Eighth international joint conference on Artificial intelligence - Volume 2
Year:
1983

Citing 3
Cited 6

On the Selection of Verbs for Natural Language Description of Traffic Scenes

GWAI '82 Proceedings of the 6th German Workshop on Artificial Intelligence
Natural language dialogue about moving objects in an automatically analyzed traffic scene

IJCAI'81 Proceedings of the 7th international joint conference on Artificial intelligence - Volume 1
Volumetric model and 3D-trajectory of a moving car derived from monocular TV-frame sequences of a street scene

IJCAI'81 Proceedings of the 7th international joint conference on Artificial intelligence - Volume 2

Generating a coherent text describing a traffic scene

COLING '86 Proceedings of the 11th coference on Computational linguistics
On the correspondence between objects and events for the diagnosis of situations in visual surveillance tasks

Pattern Recognition Letters
Knowledge and Event-Based System for Video-Surveillance Tasks

IWINAC '09 Proceedings of the 3rd International Work-Conference on The Interplay Between Natural and Artificial Computation: Part I: Methods and Models in Artificial and Natural Computation. A Homage to Professor Mira's Scientific Legacy
A task-driven intelligent workspace system to provide guidance feedback

Computer Vision and Image Understanding
Spatial knowledge representation for human-robot interaction

Spatial cognition III
Logical formalization of multimedia interpretation

Knowledge-driven multimedia information extraction and ontology evolution

Quantified Score

Hi-index	0.00

Visualization

Abstract

For an adequate interpretation of image sequences it is not only necessary to recognize objects and object positions but also certain interesting temporal developments of the scene, called events. In this paper we discuss event models for traffic scenes as high-level conceptual structures which permit interfacing to an existing natural language dialogue system. Event models are declarative descriptions of classes of events organized around verbs of locomotion. They involve components which are directly related to the deep case structure of a corresponding natural language description. Event models may be used for bottom-up scene description as well as top-down question-answering. They may also incorporate expectations about a scene, thus providing an interface to experience and common sense.