KAPLAN: A 3D Point Descriptor for Shape Completion

TL;DR

This paper introduces KAPLAN, a novel 3D point descriptor that enables efficient shape completion directly from unstructured point clouds by projecting local neighborhoods onto multiple planes and encoding their features with 2D convolutions.

Contribution

KAPLAN is a new 3D point descriptor that captures local shape information efficiently without using resource-intensive 3D convolutions, improving shape completion performance.

Findings

Achieves state-of-the-art results on public datasets

Operates directly on unstructured point clouds

Avoids expensive 3D data structures

Abstract

We present a novel 3D shape completion method that operates directly on unstructured point clouds, thus avoiding resource-intensive data structures like voxel grids. To this end, we introduce KAPLAN, a 3D point descriptor that aggregates local shape information via a series of 2D convolutions. The key idea is to project the points in a local neighborhood onto multiple planes with different orientations. In each of those planes, point properties like normals or point-to-plane distances are aggregated into a 2D grid and abstracted into a feature representation with an efficient 2D convolutional encoder. Since all planes are encoded jointly, the resulting representation nevertheless can capture their correlations and retains knowledge about the underlying 3D shape, without expensive 3D convolutions. Experiments on public datasets show that KAPLAN achieves state-of-the-art performance for 3D shape completion.