Abstract: A system and a method for training a semantic segmentation model includes obtaining a plurality of sets of images each having an index z for visibility, iteratively training the model. Iteratively training the model includes (a) for each z above 1, obtaining preliminary semantic segmentation labels for each image of the set of images of index z?1 by applying the model to each image of the set of images of index z?1, (b) processing each preliminary semantic segmentation labels using semantic segmentation labels obtained using the model on a selected image of index 1, and obtaining processed semantic segmentation labels, (c) training the model using the set of images of index z?1 and the associated processed semantic segmentation labels, and (d) performing steps (a) to (c) for z+1.

Abstract: Method of fabricating a 3-dimensional structure, mesh formwork element for fabricating a 3-dimensional structure, and method of fabricating the same. The method of fabricating a 3-dimensional structure comprises providing a mesh formwork element such that a cavity bound by at least two opposing portions of the mesh formwork is formed; accumulating a material in the cavity; and allowing the material to harden; wherein apertures in the at least two opposing portions of the mesh formwork element are adapted to the hydro-static pressure of the accumulated material or vice versa such that at least two surfaces of the hardened material substantially take on the respective shapes defined by the two opposing portions of the mesh formwork element.