Compact Local Color Descriptor Based on Rank Correlations

Abstract

In the context of object recognition, it is useful to extract, from the images, efficient local descriptors that are insensitive to the illumination conditions, to the camera scale factor and to the position and orientation of the object. In this paper, we propose to cope with this invariance problem by measuring the rank correlations between different color components of pixels located in the neighborhood of detected keypoints. This measure which takes into account both the color distribution and the spatial interactions between the pixels is stable across illumination changes. Furthermore, we show that the local descriptor based on this measure can be very compact since only 18 correlation measures are required to obtain good discriminating power. The invariance and the discriminating power of our local descriptor is assessed on a public database.