window--> split(y){~0.3:glass | 0.08:frame | ~1.22:A} A--> split(x){~0.75:B | ~0.15:glass}

Figure 1: We present an algorithm that automatically derives a split grammar for a given facade layout. A facade layout is given as a segmented facade image (left and top middle). A short subset of the generated output grammar describing a window is shown in the bottom middle. Grammars can be edited and combined to generate new stochastic split grammars. On the right we show a high-rise building that was created by such a stochastic variation. In this paper, we address the following research problem: How can we generate a meaningful split grammar that explains a given fa-cade layout? To evaluate if a grammar is meaningful, we propose a cost function based on the description length and minimize this cost using an approximate dynamic programming framework. Our eval-uation indicates that our framework extracts meaningful split gram-mars that are competitive with those of expert users, while some users and all competing automatic solutions are less successful