Spatial consistency of dense features within interest regions for efficient landmark recognition

Recently, feature grouping has been proposed as a method for improving retrieval results for logos and web images. This relies on the idea that a group of features matching over a local region in an image is more discriminative than a single feature match. In this paper, we evolve this concept further and apply it to the more challenging task of landmark recognition. We propose a novel combination of dense sampling of SIFT features with interest regions which represent the more salient parts of the image in greater detail. In place of conventional dense sampling used in category recognition that computes features on a regular grid at a number of fixed scales, we allow the sampling density and scale to vary based on the scale of the interest region. We develop new techniques for exploring stronger geometric constraints inside the feature groups and computing the match score. The spatial information is stored efficiently in an inverted index structure. The proposed approach considers part-based matching of interest regions instead of matching entire images using a histogram under bag-of-words. This helps reducing the influence of background clutter and works better under occlusion. Experiments reveal that directing more attention to the salient regions of the image and applying proposed geometric constraints helps in vastly improving recognition rates for reasonable vocabulary sizes.