Mar 21, 2023 · Here we use Graph Neural Networks (GNN) to accelerate FDL, showing that deep learning can address both limitations of FDL: it offers a 10 to 100 fold improvement in speed while also yielding...

We propose a language-independent GNN framework for document layout analysis tasks. Our proposed model, Paragraph2Graph, uses a pre-trained CNN to encode image features and incorporates 2d OCR text coordinates and image features as node features in a graph.

Apr 24, 2023 · In this paper, we present Paragraph2Graph, a language-independent graph neural network (GNN)-based model that achieves competitive results on common document layout datasets while being adaptable to business scenarios with strict separation.

Our deep neural network Graph2Plan is a learning framework for automated floorplan generation from layout graphs. The trained network can generate floorplans based on an input building boundary only (a-b), like in previous works.

Apr 27, 2020 · We introduce a learning framework for automated floorplan generation which combines generative modeling using deep neural networks and user-in-the-loop designs to enable human users to provide sparse design constraints. Such constraints are …

Jan 30, 2025 · Simple GNN Model Approach (SGNN): Based on the feature extraction network in the LFENet model, this approach uses only the raw coordinates of the graph layouts as input to perform both objective and subjective evaluation tasks.

Jul 7, 2021 · In this article, we propose a Convolutional-Graph-Neural-Network-based deep learning framework, DeepGD, which can draw arbitrary graphs once trained. It attempts to generate layouts by compromising among multiple prespecified aesthetics considering a good graph layout usually complies with multiple aesthetics simultaneously.

We present a deep neural network to predict structural similarity between 2D layouts by leveraging Graph Match-ing Networks (GMN). Our network, coined LayoutGMN, learns the layout metric via neural graph matching, using an attention-based GMN designed under a …

Here we use Graph Neural Networks (GNN) to accelerate FDL, showing that deep learning can address both limitations of FDL: it offers a 10 to 100 fold improvement in speed while also yielding layouts which are more informative.

Jun 27, 2021 · In this paper, we propose a Convolutional Graph Neural Network based deep learning framework, DeepGD, which can draw arbitrary graphs once trained. It attempts to generate layouts by compromising among multiple pre-specified aesthetics considering a good graph layout usually complies with multiple aesthetics simultaneously.