Discrete Element Texture Synthesis

1 A variety of natural and man-made phenomena can be characterized 2 by repetitive discrete elements. Examples include a stack of fruits, 3 a plate of dish, or a stone sculpture. Although certain results can 4 be produced via existing methods based on procedural or physical 5 simulation, these are often designed for specific applications. Some 6 of these methods can also be hard to control. 7 We present discrete element texture synthesis, a data-driven method 8 for placing repetitive discrete elements within a given large-scale 9 structure. Our main goal is to provide a general approach that works 10 for a variety of phenomena instead of just specific scenarios. We 11 also want it easy to use, as the user only needs to specify an in12 put exemplar for the detailed elements and the overall output struc13 ture, and our approach will produce the desired combination. Our 14 method is inspired by texture synthesis, a methodology tailored for 15 generating repetitions. However, existing texture synthesis meth16 ods cannot adequately handle discrete elements, often producing 17 unnaturally broken or merged elements. Our method not only pre18 serves the individual element properties such as color, shape, size, 19 and orientation, but also their aggregate distributions. It also allows 20 certain application specific controls, such as boundary constraints 21 for physically realistic appearance. Our key idea is a new neighbor22 hood metric for element distribution as well as an energy formu23 lation for synthesis quality and control. As an added benefit, our 24 method can also be applied for editing element distributions. 25