APC: Transferable and Efficient Adversarial Point Counterattack for Robust 3D Point Cloud Recognition

TL;DR

This paper introduces APC, a lightweight purification module that generates instance-specific counter-perturbations to defend 3D point cloud recognition models against adversarial attacks, achieving high transferability and robustness.

Contribution

APC is a novel input-level purification method that enforces geometric and semantic consistency, enabling effective and transferable defense against diverse adversarial attacks.

Findings

APC achieves state-of-the-art defense performance on benchmark datasets.

APC demonstrates superior transferability across different models.

APC operates efficiently with negligible overhead during inference.

Abstract

The advent of deep neural networks has led to remarkable progress in 3D point cloud recognition, but they remain vulnerable to adversarial attacks. Although various defense methods have been studied, they suffer from a trade-off between robustness and transferability. We propose Adversarial Point Counterattack (APC) to achieve both simultaneously. APC is a lightweight input-level purification module that generates instance-specific counter-perturbations for each point, effectively neutralizing attacks. Leveraging clean-adversarial pairs, APC enforces geometric consistency in data space and semantic consistency in feature space. To improve generalizability across diverse attacks, we adopt a hybrid training strategy using adversarial point clouds from multiple attack types. Since APC operates purely on input point clouds, it directly transfers to unseen models and defends against attacks targeting them without retraining. At inference, a single APC forward pass provides purified point clouds with negligible time and parameter overhead. Extensive experiments on two 3D recognition benchmarks demonstrate that the APC achieves state-of-the-art defense performance. Furthermore, cross-model evaluations validate its superior transferability. The code is available at https://github.com/gyjung975/APC.