Effective sketch retrieval based on its contents

Abstract

Content of sketch is different to that of image, since sketch is made up of strokes, not pixels, contains more structural and semantic information. In this paper, an effective approach for content-based sketch retrieval is proposed. With structural and ambiguous properties, the contents what sketch retrieval stressed would be the topology among constitutes of sketch, which accommodate geometry invariance. Sketches retrieval is achieved by means of similarity calculation of topological features representing the sketch content. The relevance feedback is also introduced to refine the retrieval results. Experiments prove the effectiveness and efficiency of the method in content-based sketch retrieval and user independent.