Conv-MPN: Convolutional Message Passing Neural Network for Structured Outdoor Architecture Reconstruction
This addresses structured geometry reconstruction for computer vision applications, offering a novel approach but likely incremental in the broader context of graph neural networks.
The paper tackles the problem of reconstructing outdoor buildings as planar graphs from single RGB images by proposing Conv-MPN, a convolutional message passing neural network that improves over existing neural solutions, as shown in evaluations on 2,000 buildings.
This paper proposes a novel message passing neural (MPN) architecture Conv-MPN, which reconstructs an outdoor building as a planar graph from a single RGB image. Conv-MPN is specifically designed for cases where nodes of a graph have explicit spatial embedding. In our problem, nodes correspond to building edges in an image. Conv-MPN is different from MPN in that 1) the feature associated with a node is represented as a feature volume instead of a 1D vector; and 2) convolutions encode messages instead of fully connected layers. Conv-MPN learns to select a true subset of nodes (i.e., building edges) to reconstruct a building planar graph. Our qualitative and quantitative evaluations over 2,000 buildings show that Conv-MPN makes significant improvements over the existing fully neural solutions. We believe that the paper has a potential to open a new line of graph neural network research for structured geometry reconstruction.