What is it about?
Polygon shapes are prevalent in the real world, with buildings on maps being a typical example. We propose a graph neural network designed to model and learn representations for polygon shapes. The learned embeddings can then be effectively utilized for classification tasks.
Featured Image
Photo by Scott Rodgerson on Unsplash
Why is it important?
Polygon shapes are common, yet there is a lack of efficient methods to represent such structures, limiting the application of machine learning and deep learning techniques for manipulation of these objects. By modeling polygons using graphs, our work enables the versatile use of graph neural network (GNN) models for tasks related to polygons
Read the Original
This page is a summary of: PolygonGNN: Representation Learning for Polygonal Geometries with Heterogeneous Visibility Graph, August 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3637528.3671738.
You can read the full text:
Resources
Contributors
The following have contributed to this page
