Computationally efficient image mosaicing using spanning tree representations

TL;DR: Two methods are proposed, which require less computation time by performing mosaicing in pairs of two sub-images at a time, without significant reconstruction losses, as evidenced by simulation results.

Abstract: Optimal image mosaicing has large computational complexity, that becomes prohibitive as the number of sub-images increases. Two methods are proposed, which require less computation time by performing mosaicing in pairs of two sub-images at a time, without significant reconstruction losses, as evidenced by simulation results.