Structured grid layouts are preferable in many 2D visual content creation scenarios since their structures facilitate further layout editing. Multiple geometry-based methods can effectively create structured grid layouts but require user-provided constraints or rules. Existing data-driven approaches have achieved remarkable layout generation performance, but fail to produce appropriate layout structures. We present GTLayout, a novel generative model for structured grid layout generation. We adopt general trees to represent structured grid layouts and exploit a recursive neural network (RvNN) for this generation task. Our model can handle grid layouts with varied structures and regular arrangements. Qualitative and quantitative experiments on public grid layout datasets show that our method outperforms several baselines in the tasks of layout reconstruction and layout generation, especially for datasets containing few samples. We also demonstrate that the structured layout space constructed by our method can blend structures of layouts, as well as providing a visualization and analysis of the layout space. Additionally, we consider two application cases based on GTLayout : multiple layout interpolation and conditional layout generation. Our code is available at https://github.com/Warren-swr/GT-Layout.

We present a novel method for blending hierarchical layouts with semantic labels. The core of our method is a hierarchical structure correspondence algorithm, which recursively finds optimal substructure correspondences, achieving a globally optimal correspondence between a pair of hierarchical layouts. This correspondence is consistent with the structures of both layouts, allowing us to define the union of the layouts' structures. The resulting compound structure helps extract intermediate layout structures, from which blended layouts can be generated via an optimization approach. The correspondence also defines a similarity measure between layouts in a hierarchically structured view. Our method provides a new way for novel layout creation. The introduced structural similarity measure regularizes the layouts in a hyperspace. We demonstrate two applications in this paper, i.e., exploratory design of novel layouts and sketch-based layout retrieval, and test them on a magazine layout dataset. The effectiveness and feasibility of these two applications are confirmed by the user feedback and the extensive results. The code is available at https://github.com/lyf7115/LayoutBlending.

Bandwidth defragmentation, i.e., the operation to reconfigure existing connections for making the spectrum usage less fragmented and less misaligned, has recently been recognized as one of the most important features for elastic optical networks (EONs). In this paper, we propose a novel comprehensive bandwidth defragmentation algorithm that considers the problems of 1) When to defragment? 2) What for defragment? and 3) How to defragment? jointly. The proposed algorithm accomplishes defragmentation through proactive network reconfiguration that only reroutes a portion of existing connections. In each defragmentation operation, we first choose the connections to reroute using a selection strategy, then determine how to reroute them with the defragmentation based routing and spectrum assignment (DF-RSA), and finally perform rerouting with best-effort traffic migration to minimize traffic disruptions. Simulation results indicate that in order to make the bandwidth blocking probability (BBP) comparable with that from the Greenfield scenario (100% rerouting all the time), the proposed algorithm only needs to reroute ~30% existing connections. The simulations also demonstrate that the traffic disruption percentages are less than 1% for defragmentations with 30% rerouting and can be further reduced to within 0.25% by adding a move-to-vacancy (MTV) approach in the traffic migration.

In elastic optical networks (EONs), bandwidth fragmentation refers to the existence of non-aligned, isolated and small-sized blocks of contiguous subcarrier slots in the optical spectrum. As they are neither contiguous in the spectrum domain nor aligned along the routing paths, the network operator will have difficulty to use these slots for future connections. In this work, we analyze the effect of bandwidth fragmentation on the blocking probability in EONs. Our theoretical analysis indicates that two factors related to bandwidth fragmentation have effects on the blocking probability: 1) the extent that the available slot-blocks (i.e., blocks of contiguous slots) on different links are aligned on spectrum locations, and 2) the sizes of the available slot-blocks in links' spectra for future requests. When an EON's spectrum becomes more fragmented, the first factor actually reduces the blocking probability, while the second one increases the blocking probability. Their mixed effect determines the overall trend of how the blocking probability will change with bandwidth fragmentation. Our theoretical model can forecast this trend and reveal the relation among the blocking probability, bandwidth fragmentation, request bandwidth distribution, and spectrum utilization. We have also conducted numerical simulations to verify the theoretical analysis, and the simulation results exhibit similar trends as predicted by the theoretical model.