Interpolated Convolutional Networks for 3D Point Cloud Understanding
This addresses the problem of 3D point cloud understanding for computer vision applications, offering a novel method for a known bottleneck in handling irregular data.
The paper tackles the challenge of applying convolutions to sparse, irregular 3D point clouds by proposing InterpConv, a novel convolution operation that interpolates features to kernel coordinates with a normalization term for sparsity invariance. It achieves state-of-the-art performance on benchmarks like ModelNet40, ShapeNet Parts, and S3DIS for tasks such as shape classification and segmentation.
Point cloud is an important type of 3D representation. However, directly applying convolutions on point clouds is challenging due to the sparse, irregular and unordered data structure. In this paper, we propose a novel Interpolated Convolution operation, InterpConv, to tackle the point cloud feature learning and understanding problem. The key idea is to utilize a set of discrete kernel weights and interpolate point features to neighboring kernel-weight coordinates by an interpolation function for convolution. A normalization term is introduced to handle neighborhoods of different sparsity levels. Our InterpConv is shown to be permutation and sparsity invariant, and can directly handle irregular inputs. We further design Interpolated Convolutional Neural Networks (InterpCNNs) based on InterpConv layers to handle point cloud recognition tasks including shape classification, object part segmentation and indoor scene semantic parsing. Experiments show that the networks can capture both fine-grained local structures and global shape context information effectively. The proposed approach achieves state-of-the-art performance on public benchmarks including ModelNet40, ShapeNet Parts and S3DIS.