Non-geometric Detection and Non-symbolic Visualization of Maneuvering Affordance

K. Kamejima (Japan)

Keywords

Maneuvering Affordance; Similarity Rule; Directional Scale Image; Self-Similarity; Multi-Scale Image; Com puter Vision; Signal and Image Processing

Abstract

For detecting and visualizing maneuvering affordance, scale information is organized within an intentional coordinate system. By considering the mental dynam ics as a self-similarity process, the maneuvering affor dance in complex scene is represented by generalized mental dynamics. The brightness distribution of the directional scale image is non-geometricaly identified with fractal attractors. Capturing probability for the attractors is generated to visualize the expansion of the affordance without explicit symbol reference. The detectability of affordance patterns has been verified through experimental studies.

Important Links:



Go Back