Method and system to generate a 3D representation in a dynamically changing 3D scene

摘要

Method for filling the gaps (802) remaining in the bottom of an image (201) after trimming the bottom objects, comprising the steps of - acquiring (102) an image (201) of a camera (702) that observes a 3D scene (701); - segmenting the image (201) by separating the objects from the background of the image (201); - filling the corresponding gaps (802) remaining in the background by restoring the image, mapping the image data corresponding to the original patches (804a, 804b, 804c, 804d) that comprise real image data to the patches of destination (803a, 803b, 803c, 803d) comprising unknown portions of the image, thus filling the gaps (802); wherein the mapping involves a transformation of the patches according to their spatial relationship, that is, 3D characterized in that an unknown part of the image is known to comprise a mark element (807), in which a mark element is an immobile background element whose nominal location is known and is a linearly extending element on a playing field, and the method comprises the steps of selecting a destination patch (803c, 803d) to cover at least part of the mark (807), and search the image for a corresponding origin patch (804c, 804d) along the mark (806a).