A method, apparatus and computer program product provide for encoding and decoding volumetric video data. In relation to encoding, one or more patches into which a point cloud representative of volumetric video data has been decomposed are evaluated by identifying one or more areas within a respective patch in which a difference in depth values between corresponding pixels of at least two depth planes satisfy a predefined threshold. For an area within the respective patch that has been identified, a sub patch is generated representative of the depth values of one of the depth planes. The patches are mapped including the one or more patches and the sub patch to a grid and an image is generated based upon the patches mapped to the grid.