An encoder for encoding a video signal, comprising a video modelling module configured for determining a plurality of video modelling parameters for a plurality of video locations on a spatiotemporal grid of said video signal, said spatiotemporal grid comprising at least two spatial dimensions and a time dimension, each video modelling parameter being adapted for allowing a pre-determined video model to at least approximately reconstruct its video location, a video segmentation module configured for segmenting said video signal into a plurality of spatiotemporal video regions, based on said video modelling parameters and a vectorisation module configured for vectorising spatiotemporal surfaces of said spatiotemporal video regions; and wherein said encoder is configured for encoding said video signal based on at least a subset of said determined plurality of video modelling parameters and based on said vectorised spatiotemporal surfaces, wherein said subset is determined taking into account said spatiotemporal video regions.