We propose an interactive sketching tool called SKIT to explore image database. The aim is to achieve fast result convergence according to the visual user query. Our main contribution is a new interactive image exploration approach which dynamically adapts to user sketches and provides feedback. The novel user interface is suitable for a range of interactive image-database access applications. In addition, we propose a position-aware matching approach for SKIT to support translation-free sketch searching. Experimental results demonstrate that our method outperforms state-of-the-art approaches with respect to the superior user interface and matching results.
CITATION STYLE
Hu, S., Zhang, H., Zhang, S., Fang, Z., & Huang, Q. (2016). Sketch-based retrieval in large-scale image database via position-aware Silhouette Matching. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9654, pp. 243–256). Springer Verlag. https://doi.org/10.1007/978-3-319-40259-8_22
Mendeley helps you to discover research relevant for your work.