Deep Unsupervised Learning of 3D Point Clouds via Graph Topology Inference and Filtering
arXiv:1905.04571 · doi:10.1109/TIP.2019.2957935
Abstract
We propose a deep autoencoder with graph topology inference and filtering to achieve compact representations of unorganized 3D point clouds in an unsupervised manner. Many previous works discretize 3D points to voxels and then use lattice-based methods to process and learn 3D spatial information; however, this leads to inevitable discretization errors. In this work, we handle raw 3D points without such compromise. The proposed networks follow the autoencoder framework with a focus on designing the decoder. The encoder adopts similar architectures as in PointNet. The decoder involves three novel modules. The folding module folds a canonical 2D lattice to the underlying surface of a 3D point cloud, achieving coarse reconstruction; the graph-topology-inference module learns a graph topology to represent pairwise relationships between 3D points, pushing the latent code to preserve both coordinates and pairwise relationships of points in 3D point clouds; and the graph-filtering module couples the above two modules, refining the coarse reconstruction through a learnt graph topology to obtain the final reconstruction. The proposed decoder leverages a learnable graph topology to push the codeword to preserve representative features and further improve the unsupervised-learning performance. We further provide theoretical analyses of the proposed architecture. In the experiments, we validate the proposed networks in three tasks, including 3D point cloud reconstruction, visualization, and transfer classification. The experimental results show that (1) the proposed networks outperform the state-of-the-art methods in various tasks; (2) a graph topology can be inferred as auxiliary information without specific supervision on graph topology inference; and (3) graph filtering refines the reconstruction, leading to better performances.
arXiv commentTo appear in IEEE Transactions on Image Processing
References in corpus (12)
- The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
- Geometric deep learning: going beyond Euclidean data
- Discrete Signal Processing on Graphs: Sampling Theory
- Learning graphs from data: A signal representation perspective
- Perfect Reconstruction Two-Channel Wavelet Filter-Banks for Graph Structured Data
- Efficient Sampling Set Selection for Bandlimited Graph Signals Using Graph Spectral Proxies
- Deep Parametric Continuous Convolutional Neural Networks
- Convolutional Neural Network Architectures for Signals Supported on Graphs
- Signal Recovery on Graphs: Variation Minimization
- Graph-based compression of dynamic 3D point cloud sequences
- Fast Resampling of 3D Point Clouds via Graphs
- Subgraph-based filterbanks for graph signals