Query-by-sketch image retrieval using data mining for bridging semantic gap

点击翻译按钮获取中文摘要

The purpose of this study is to bridge the semantic gap between image feature and human subjective using data-mining from huge image data, sketch images and user evaluations and to improve the effectiveness and efficiency of interactive query-by-sketch image retrieval. The query-by-sketch image retrieval using sketch prediction is focused on online stroke. Sketch prediction enables users to obtain target retrieval images without completely finishing input sketch and to bridge the semantic gap.