3DVerifier: efficient robustness verification for 3D point cloud models

Ronghui Mu

doi:10.1007/s10994-022-06235-3

2022-11-02

Ronghui Mu

Wenjie Ruan *

Leandro Soriano Marcolino

Qiang Ni

DOI：10.1007/s10994-022-06235-3

原文链接

原文求助

摘要

En 中文

3D point cloud models are widely applied in safety-critical scenes, which delivers an urgent need to obtain more solid proofs to verify the robustness of models. Existing verification method for point cloud model is time-expensive and computationally unattainable on large networks. Additionally, they cannot handle the complete PointNet model with joint alignment network that contains multiplication layers, which effectively boosts the performance of 3D models. This motivates us to design a more efficient and general framework to verify various architectures of point cloud models. The key challenges in verifying the large-scale complete PointNet models are addressed as dealing with the cross-non-linearity operations in the multiplication layers and the high computational complexity of high-dimensional point cloud inputs and added layers. Thus, we propose an efficient verification framework, 3DVerifier, to tackle both challenges by adopting a linear relaxation function to bound the multiplication layer and combining forward and backward propagation to compute the certified bounds of the outputs of the point cloud models. Our comprehensive experiments demonstrate that 3DVerifier outperforms existing verification algorithms for 3D models in terms of both efficiency and accuracy. Notably, our approach achieves an orders-of-magnitude improvement in verification efficiency for the large network, and the obtained certified bounds are also significantly tighter than the state-of-the-art verifiers. We release our tool 3DVerifier via for use by the community.

Keyword:

Robustness verification

3D point cloud models

Adversarial attacks

Deep neural networks

PointNet