Survey on Sketch Based Image Retrieval

Abstract: Sketch-based communication is the oldest form of writing. In which sketch depicts rough shape of object. Sketch-based image retrieval (SBIR) can therefore be a very valuable information search tool. Although sketch is good way to express people’s thoughts, there is a large gap in the appearance of user sketches and photorealistic images, when people sketch, they usually focus on the main structure of an object and only draw the semantic contour boundary. In contrast, photo-realistic images contain the color, texture and detailed shape of an object, which makes it very difficult to directly match a sketch and the corresponding photo-realistic image. Therefore, this is fundamental challenge in SBIR. The existence of noisy edges on photo realistic image degrades retrieval performance and to bridge this gap there is framework consisting of line segment descriptor named and noise impact reduction algorithm. Descriptor extracts edges and captures the relationship between them. Object boundary selection algorithm used to reduce the impact of noisy edges. The hypothesis is used to maximize retrieval score, for which multiple hypotheses are generated. In scoring process there are sometimes false matches happens, to reduce such distraction, two constraints on spatial and coherent aspects are used.

Keywords: Descriptor, sketch retrieval, edge based, histogram, line relationship.