Parameter is Not All You Need: Starting from Non-Parametric Networks for 3D Point Cloud Analysis

TL;DR

This paper introduces a non-parametric network for 3D point cloud analysis that performs well without training, and extends it to parametric models and as a plug-in module to improve existing methods.

Contribution

It presents a purely non-learnable 3D point cloud analysis method and demonstrates its use as a foundation for parametric models and as a plug-in for existing models.

Findings

Non-parametric Point-NN performs well on 3D tasks without training.

Point-NN can be extended to parametric networks with high performance-efficiency.

Point-NN enhances existing 3D models during inference without re-training.

Abstract

We present a Non-parametric Network for 3D point cloud analysis, Point-NN, which consists of purely non-learnable components: farthest point sampling (FPS), k-nearest neighbors (k-NN), and pooling operations, with trigonometric functions. Surprisingly, it performs well on various 3D tasks, requiring no parameters or training, and even surpasses existing fully trained models. Starting from this basic non-parametric model, we propose two extensions. First, Point-NN can serve as a base architectural framework to construct Parametric Networks by simply inserting linear layers on top. Given the superior non-parametric foundation, the derived Point-PN exhibits a high performance-efficiency trade-off with only a few learnable parameters. Second, Point-NN can be regarded as a plug-and-play module for the already trained 3D models during inference. Point-NN captures the complementary geometric knowledge and enhances existing methods for different 3D benchmarks without re-training. We hope our work may cast a light on the community for understanding 3D point clouds with non-parametric methods. Code is available at https://github.com/ZrrSkywalker/Point-NN.