Human activity recognition from object interaction in domestic scenarios

Abstract

This paper presents a real time approach to the recognition of human activity based on the interaction between people and objects in domestic settings, specifically in a kitchen. Regarding the procedure, it is based on capturing partial images where the activity takes place using a colour camera, and processing the images to recognize the present objects and their location. For object description and recognition, a histogram on rg chromaticity space has been selected. The interaction with the objects is classified into four types of possible actions; (unchanged, add, remove or move). Activities are defined as recipes, where objects play the role of ingredients, tools or substitutes. Sensed objects and actions are then used to analyze in real time the probability of the human activity performed at a particular moment in a continuous activity sequence.