Robust 3D Point Clouds Classification based on Declarative Defenders

TL;DR

This paper proposes novel algorithms for mapping 3D point clouds into 2D images to improve classification accuracy and robustness, leveraging foundation models and analyzing defense mechanisms against adversarial attacks.

Contribution

It introduces three new algorithms for 3D to 2D mapping that reduce domain gap and enhance robustness in point cloud classification.

Findings

Proposed algorithms outperform existing methods in accuracy.

Enhanced robustness against adversarial attacks.

Regular 2D images from generative models improve domain transfer.

Abstract

3D point cloud classification requires distinct models from 2D image classification due to the divergent characteristics of the respective input data. While 3D point clouds are unstructured and sparse, 2D images are structured and dense. Bridging the domain gap between these two data types is a non-trivial challenge to enable model interchangeability. Recent research using Lattice Point Classifier (LPC) highlights the feasibility of cross-domain applicability. However, the lattice projection operation in LPC generates 2D images with disconnected projected pixels. In this paper, we explore three distinct algorithms for mapping 3D point clouds into 2D images. Through extensive experiments, we thoroughly examine and analyze their performance and defense mechanisms. Leveraging current large foundation models, we scrutinize the feature disparities between regular 2D images and projected 2D images. The proposed approaches demonstrate superior accuracy and robustness against adversarial attacks. The generative model-based mapping algorithms yield regular 2D images, further minimizing the domain gap from regular 2D classification tasks. The source code is available at https://github.com/KaidongLi/pytorch-LatticePointClassifier.git.