Spatial Segmentation of Human Activities in Intelligent Space

Abstract

This paper presents approaches using the spatial memory to classify human activities spatially and characteristically for recognizing space-human activity association. The spatial memory enables humans to store computerized information such as digital files and commands into the real world by assigning three-dimensional position as the memory addresses. Humans can retrieve and store such information by directly indicating the point using their own body e.g. user's hand and user's head. A virtual tag, which associates a digital data with a spatial location, is called Spatial-Knowledge-Tag (SKT). SKT is created by a human based on his policy and his purpose in an environment. SKTs including locations and contents describe human activities at places. To design and implement services for inhabitants in an intelligent environment, it is required to know purposes which people intend to use the places. This paper applies two approaches to classify human activities spatially and characteristically. More specifically, one is a classification of arranged SKTs by using only spatial memory addresses and the size of accessible regions, the other is a classification of arranged SKT based on observation of human activities and forms of spatial memory data.