Particle swarm over scene matching

Advances in a scene matching approach based on objects spatial relationships are discussed. Objects spatial relationships are captured using force histograms, which are used as affine-invariant image descriptors. Object mapping across images is performed by finding the best correspondence map between image descriptors. Due to a very large search space, the particle swarm optimization method is used to facilitate more efficient search. The image descriptors map is then translated into an object map identifying the correspondences between objects in the two images.