Increasing depth lateral resolution based on sensor fusion

Technologies for measuring range data were intensively developed in recent years. The Time-Of-Flight matrix range sensor is one of the most active range acquisition devices, due to its effective suppression of background illumination, relative low cost and high portability. However, the resolution of such 3D range sensor is much lower than that of a modern colour image. This paper presents a solution to the problem of deriving a high-resolution depth image from a low-resolution sensor and a high-resolution colour image. The proposed algorithm is a modification of the Markov Random Fields fusion method that was presented by Diebel and Thrun (2005). Two extensions are suggested to improve the optimisation criteria.