PointACL:Adversarial Contrastive Learning for Robust Point Clouds Representation under Adversarial Attack

TL;DR

PointACL introduces a novel adversarial contrastive learning approach for 3D point cloud representations, enhancing robustness against adversarial attacks by utilizing unprojected features and high-difference points, validated on classification and segmentation tasks.

Contribution

The paper proposes a new method for generating high-quality 3D adversarial examples using unprojected features and a robust aware loss, improving adversarial robustness in contrastive learning.

Findings

Achieves comparable robust accuracy to state-of-the-art methods.

Utilizing high-difference points significantly boosts robustness.

Effective in 3D classification and segmentation tasks.

Abstract

Despite recent success of self-supervised based contrastive learning model for 3D point clouds representation, the adversarial robustness of such pre-trained models raised concerns. Adversarial contrastive learning (ACL) is considered an effective way to improve the robustness of pre-trained models. In contrastive learning, the projector is considered an effective component for removing unnecessary feature information during contrastive pretraining and most ACL works also use contrastive loss with projected feature representations to generate adversarial examples in pretraining, while "unprojected " feature representations are used in generating adversarial inputs during inference.Because of the distribution gap between projected and "unprojected" features, their models are constrained of obtaining robust feature representations for downstream tasks. We introduce a new method to generate high-quality 3D adversarial examples for adversarial training by utilizing virtual adversarial loss with "unprojected" feature representations in contrastive learning framework. We present our robust aware loss function to train self-supervised contrastive learning framework adversarially. Furthermore, we find selecting high difference points with the Difference of Normal (DoN) operator as additional input for adversarial self-supervised contrastive learning can significantly improve the adversarial robustness of the pre-trained model. We validate our method, PointACL on downstream tasks, including 3D classification and 3D segmentation with multiple datasets. It obtains comparable robust accuracy over state-of-the-art contrastive adversarial learning methods.