A method of generating an image comprises receiving (301, 303) a first and second texture map and mesh representing a scene from a first view point and second view point respectively. A light intensity image is generated (305) for a third view point. For a first position this includes determining (401, 403) a first and second light intensity value for the first position by a view point transformation based on the first texture map and the first mesh and on the second texture map and the second mesh respectively. The light intensity value is then determined (405) by a weighted combination of the first and second light intensity values. The weighting depends on a depth gradient in the first mesh at a first mesh position corresponding to the first position relative to a depth gradient in the second mesh at a second mesh position corresponding to the first position.