Development of an Intelligent Fabric Retrieval System using Com- puter-Based Kansei Algorithm

The objective of this research is to develop an intelligent fabric retrieval system using computer-based Kansei algorithms to assist fashion designers in designing costume textiles, as well as to help consumers find their preferred textile samples quickly and efficiently. The fabric retrieval system, based on human psychological Kansei and perception, can stimulate and increase the creativity of textile designers and other visual artifact designers alike. The fabric retrieval system is able to assist designers in creating new ideas through two ways interactively. The first is the high-level Kansei searching algorithm, where predefined impression words are systematically adjusted. The other is the low-level perception query, which includes perception feature modification and image similarity indexing. Moreover, the two ways are correlated with each other through a Neural Network. That is, the change of the values of low-level perception features, by users, could affect those of the Kansei features and the vision features of images simultaneously. Subsequently users could retrieve similar samples based on the impression words or vision features in this content-based retrieval system with a fabric database.