C-FA: Enhancing Transferability through Contrastive Learning and Feature Attention in Point Cloud Attacks
- 2 February 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-14
Abstract
3D classification networks are vulnerable to adversarial examples, which can cross decision boundaries and cause the model to misclassify. However, the adversarial examples generated by the current method do not transferred well across the decision boundaries of various models, resulting in a lower success rate of transfer to a black-box model and poor attack performance against the target model. In this paper, we propose a new method, called C-FA, which revisits the problem of crossing model decision boundaries, generating adversarial examples with improved transferability. We use a contrastive learning strategy, by increasing the difference between the adversarial examples and the benign examples in the feature space, prompting the adversarial examples to cross the decision boundary, leading to misclassification by the model. In addition, considering that perturbing the salient features focused on by the model can effectively enhance the transferability of the generated adversarial examples, we introduce a feature attention mechanism that identifies and weights features that have a more significant impact on decision-making by analyzing the feature gradient. The effectiveness of C-FA and its competitiveness against existing state-of the-art techniques are confirmed through extensive experiments on six representative networks including PointNet, PointNet++, DGCNN, PointConv, PCT, and Mamba3D.Keywords
Funding Information
- National Natural Science Foundation of China (62476001)
- Higher Education Research Program Project of Anhui Province (2024AH010004)
- Scientific Research Planning Project of Anhui Province (2022AH050072)
This publication has 73 references indexed in Scilit:
- SaliencyI2PLoc: Saliency-guided image–point cloud localization using contrastive learningInformation Fusion, 2025
- Exploring the Devil in Graph Spectral Domain for 3D Point Cloud Attacks2022
- 3D Object Detection for Autonomous Driving: A SurveyPattern Recognition, 2022
- Boosting 3D Adversarial Attacks With Attacking on FrequencyIEEE Access, 2022
- A novel transferability attention neural network model for EEG emotion recognitionNeurocomputing, 2021
- PCT: Point cloud transformerComputational Visual Media, 2021
- FineFool: A novel DNN object contour attack on image recognition based on the attention perturbation adversarial techniqueComputers & Security, 2021
- PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding2020
- AdvPC: Transferable Adversarial Perturbations on 3D Point Clouds2020
- Transferable Adversarial Perturbations2018