A Kansei Image Retrieval Based on Emotion Model

王上飞,陈恩红,王胜惠,王煦法
DOI: https://doi.org/10.3969/j.issn.1007-0249.2003.06.011
2003-01-01
Abstract:The target of Kansei information processing is to simulate and recognize human sensibility, sensuality or emotion, to realize natural and harmonious human-machine interface as to achieve human-centered world. People usually use adjectives to express emotion. An emotional model based on adjectives is provided in this paper. Based on the idea of dimension?from psychology, semantic differential method and factor analysis are used to construct an orthogonal emotional space. The measurement of similarity in the emotional space is analysis. After that, the emotion model is used in image retrieval. Color and shape features extracted from the image can be used as the kansei features. A radial basis function neural network is used for mapping each image from the feature space to the emotional space. Thus an kansei image retrieval system is obtained. At last an interesting experimental result is presented.
What problem does this paper attempt to address?