Semantic human image classification based on Energy-Action model

Abstract

The classification of semantic human images is an active problem in interpreting multimedia images. Many researchers have attempted to improve the semantic model by using semantic action concepts. Although previous techniques, such as keyword definition and using content features of human action, have been applied, most results indicate that human images cannot be mapped into actual image meaning. The aim of this paper is to classify semantic human images by integrating relevant image contents and the energy expenditure. We proposed a new semantic model called the energy-action (EA) model, which analyzes the energy intensity of body parts with essential reference points. The EA model is presented in two stages: human action analysis and energy intensity analysis. Our results indicate that the proposed approach offers significant performance improvements in the interpretation of semantic human images.