Image Database System of Clothing Emotion

Zhang, Haibo,Tiejun Huang,Liu, Ruipu
DOI: https://doi.org/10.1109/icebeg.2011.5882605
2011-01-01
Abstract:In the fields of apparel and artificial intelligence, clothing images, which are closely related to human life, have important theoretical and practical significance when they are used to assess, understand and analyze the clothing emotion of the image subjects. At present, the study of clothing emotion still stays at the stage of qualitative description, so it cannot meet the needs for people to do further study in both the above fields. In this paper, by taking the male western-style clothing as an example, we have established a clothing emotion image database, through gathering, selecting, and voting scoring the dress emotion vocabulary. With this database, we can determine the group of adjectives which can best describe people's emotion for men suits, realizing the quantitative description of clothing emotion words which can in turn be used to retrieve clothing images. Therefore, our study has laid a foundation for clothing evaluation as well as for clothing image pattern recognition.
What problem does this paper attempt to address?