Differentiable Convolution Search for Point Cloud Processing

TL;DR

This paper introduces PointSeaConv, a data-driven, differentiable convolution search method for point clouds, enabling automatic design of effective geometric shape modeling networks.

Contribution

It proposes a novel neural architecture search paradigm for point cloud convolution, including joint optimization and an epsilon-greedy algorithm to improve shape modeling.

Findings

PointSeaNet outperforms handcrafted models on benchmarks.

The method effectively captures geometric shapes.

Joint optimization improves network performance.

Abstract

Exploiting convolutional neural networks for point cloud processing is quite challenging, due to the inherent irregular distribution and discrete shape representation of point clouds. To address these problems, many handcrafted convolution variants have sprung up in recent years. Though with elaborate design, these variants could be far from optimal in sufficiently capturing diverse shapes formed by discrete points. In this paper, we propose PointSeaConv, i.e., a novel differential convolution search paradigm on point clouds. It can work in a purely data-driven manner and thus is capable of auto-creating a group of suitable convolutions for geometric shape modeling. We also propose a joint optimization framework for simultaneous search of internal convolution and external architecture, and introduce epsilon-greedy algorithm to alleviate the effect of discretization error. As a result, PointSeaNet, a deep network that is sufficient to capture geometric shapes at both convolution level and architecture level, can be searched out for point cloud processing. Extensive experiments strongly evidence that our proposed PointSeaNet surpasses current handcrafted deep models on challenging benchmarks across multiple tasks with remarkable margins.