Logo image
Back
PD-Net: Point Dropping Network for Flexible Adversarial Example Generation with Lo Regularization
Conference paper

PD-Net: Point Dropping Network for Flexible Adversarial Example Generation with Lo Regularization

Z. Wang, X. Wang, F. Sohel and Y. Liao
2021 International Joint Conference on Neural Networks (IJCNN)
International Joint Conference on Neural Networks (IJCNN) 2021 (Shenzhen, China, 18/07/2021–22/07/2021)
2021
DOI: https://doi.org/10.1109/IJCNN52387.2021.9533701
url
Link to Published Version *Subscription may be requiredView

Abstract

It is a challenging task to generate adversarial point clouds, considering the irregular structure of a point cloud, the large search space, and the requirement of imperception to humans. In this paper, a flexible adversarial point cloud generation method, named Point Dropping Network (PD-Net), is proposed, which can be trained to craft adversarial examples in a single forward pass. The network is designed to launch untargeted black-box attacks to deep 3D models through point dropping regularized by the L0 norm, in contrast to the widely adopted point perturbation methods. To enable incorporation into a deep neural network, the probability of a point to be dropped, which can be described by a Bernoulli distribution, is approximated by a hard concrete distribution. The network of PD-Net consists of an encoder and a decoder, where the former encodes geometric information of each point and the latter learns to drop points from their local features in an unsupervised way. Experiments on two popular deep 3D models (including PointNet and PointNet++) show that the proposed PD-Net degrades the recognition accuracy to a large extent and achieves a high flexibility at the same time.

Details

Metrics

100 Record Views
Logo image