iRP is a contour-based image retrieval prototype system for the large scale database. By simulating hierarchical image processing of human visual system, it is able to retrieve relative images from large scale database according to the contour of the object in an image. In most of situation, users usually want to find images with certain contours without caring its size or location in the image. But most existing algorithms are not able to deal with such invariance problems. The existing retrieval systems usually require that the objects in the query image and in the retrieved image should have the same position and almost the same size. Different from existing algorithms, iRP is developed to address these invariance issues, such as the shift-invariance and scale-invariance problem. The system could perform an image search from 4 million images quickly, and the response time for each query is about 3~5 seconds.
The system provides a panel to the users to draw a sketch whatever they imagine, and it will search the database quickly by the sketch and return the retrieval results. If the users aren’t good at drawing and the sketch doesn’t like what they imagine very much, the system still allows the other query method: a natural image, so the users could refine the results by choosing one natural image as the new query from the list of previous sketch retrieval results. It brings the better interactive user experiment and the convenience for those who don't well in drawing.
|
