FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis
This addresses the challenge of applying CNNs to non-grid data for 3D shape analysis, offering a novel method that improves shape correspondence without relying on predefined descriptors.
The paper tackled the problem of extending convolutional neural networks to irregular graph-structured data like 3D shape meshes by proposing a novel graph-convolution operator that dynamically computes correspondences from learned features, resulting in significant improvements over previous state-of-the-art shape correspondence results.
Convolutional neural networks (CNNs) have massively impacted visual recognition in 2D images, and are now ubiquitous in state-of-the-art approaches. CNNs do not easily extend, however, to data that are not represented by regular grids, such as 3D shape meshes or other graph-structured data, to which traditional local convolution operators do not directly apply. To address this problem, we propose a novel graph-convolution operator to establish correspondences between filter weights and graph neighborhoods with arbitrary connectivity. The key novelty of our approach is that these correspondences are dynamically computed from features learned by the network, rather than relying on predefined static coordinates over the graph as in previous work. We obtain excellent experimental results that significantly improve over previous state-of-the-art shape correspondence results. This shows that our approach can learn effective shape representations from raw input coordinates, without relying on shape descriptors.