GASCN: Graph Attention Shape Completion Network
This addresses shape completion for robotics and computer vision, presenting an incremental improvement over existing methods.
The paper tackles the problem of inferring complete object geometry from partial point clouds by proposing GASCN, a neural network model that combines graph-based local encoding with MLP-based global encoding to produce dense and precise shape completions, outperforming standard methods on the Shapenet benchmark.
Shape completion, the problem of inferring the complete geometry of an object given a partial point cloud, is an important problem in robotics and computer vision. This paper proposes the Graph Attention Shape Completion Network (GASCN), a novel neural network model that solves this problem. This model combines a graph-based model for encoding local point cloud information with an MLP-based architecture for encoding global information. For each completed point, our model infers the normal and extent of the local surface patch which is used to produce dense yet precise shape completions. We report experiments that demonstrate that GASCN outperforms standard shape completion methods on a standard benchmark drawn from the Shapenet dataset.