Behavioral Knowledge Representation for the Understanding and Creation of Video Sequences

The algorithmic generation of textual descriptions of real world image sequences requires conceptual knowledge. The algorithmic generation of synthetic image sequences from textual descriptions requires conceptual knowledge, too. An explicit representation formalism for behavioral knowledge based on formal logic is presented which can be utilized in both tasks – Understanding and Creation of video sequences. Common sense knowledge is represented at various abstraction levels in a Situation Graph Tree. This form of representation is exploited in order to fill in missing details in a natural language text describing developments for an image sequence to be synthesized.

[1]  Karl Schäfer,et al.  Unscharfe zeitlogische Modellierung von Situationen und Handlungen in Bildfolgenauswertung und Robotik , 1996, DISKI.

[2]  Hans-Hellmut Nagel,et al.  Image sequence evaluation: 30 years and still going strong , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[3]  Uwe Reyle,et al.  From discourse to logic , 1993 .

[4]  Arjan Egges,et al.  Generating a 3D simulation of a car accident from a formal description , 2001 .

[5]  Hans-Hellmut Nagel,et al.  Representation of Behavioral Knowledge for Planning and Plan-Recognition in a Cognitive Vision System , 2002, KI.

[6]  Amitabha Mukerjee,et al.  Visualisation of Conceptual Descriptions Derived from Image Sequences , 1999, DAGM-Symposium.

[7]  Gerhard Lakemeyer,et al.  KI 2002: Advances in Artificial Intelligence , 2002, Lecture Notes in Computer Science.

[8]  Amitabha Mukerjee,et al.  Conceptual description of visual scenes from linguistic models , 2000, Image Vis. Comput..

[9]  Jörg R. J. Schirra,et al.  Bildbeschreibung als Verbindung von visuellem und sprachlichem Raum - eine interdisziplinäre Untersuchung von Bildvorstellungen in einem Hörermodell , 1994, DISKI.

[10]  Hans-Hellmut Nagel,et al.  ‘Occurrence’ Extraction from Image Sequences of Road Traffic Scenes , 2002 .

[11]  Hans-Hellmut Nagel,et al.  Natural Language Texts for a Cognitive Vision System , 2002, ECAI.

[12]  Anthony G. Cohn,et al.  Constructing qualitative event models automatically from video input , 2000, Image Vis. Comput..

[13]  Ralf Gerber,et al.  Natürlichsprachliche Beschreibung von Straáenverkehrsszenen durch Bildfolgenauswertung [online] , 2000 .

[14]  Hans-Hellmut Nagel,et al.  Incremental recognition of traffic situations from video image sequences , 2000, Image Vis. Comput..

[15]  Ralf Gerber.  Natürlichsprachliche Beschreibung von Straßenverkehrsszenen durch Bildfolgenauswertung , 1999 .

[16]  François Brémond,et al.  Human Behaviour Visualisation and Simulation for Automatic Video Understanding , 2002, WSCG.

[17]  I ap Gwynn,et al.  Differentiation of rat osteoblast-like cells in monolayer and micromass cultures. , 2002, European cells & materials.

[18]  Richard Sproat,et al.  WordsEye: an automatic text-to-scene conversion system , 2001, SIGGRAPH.

[19]  Hilary Buxton,et al.  Conceptual descriptions from monitoring and watching image sequences , 2000, Image Vis. Comput..

[20]  Jörg R. J. Schirra,et al.  Optional Deep Case Filling and Focus Control with Mental Images: ANTLIMA-KOREF , 1995, IJCAI.