Shape Prior Guided Attack: Sparser Perturbations on 3D Point Clouds
Authors: Zhenbo Shi, Zhi Chen, Zhenbo Xu, Wei Yang, Zhidong Yu, Liusheng Huang8277-8285
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments show that our method has a higher success rate (even under defense), higher transferability and less time consumption. We test the performance of SPGA on a much broader range of point cloud classifiers, which are more comprehensive than previous attacks and are particularly persuasive against transferability. We use two public datasets, 3D MNIST and the aligned Model Net40 (Wu et al. 2015). |
| Researcher Affiliation | Academia | Zhenbo Shi1, Zhi Chen1, Zhenbo Xu2, Wei Yang1*, Zhidong Yu1 Liusheng Huang1 1University of Science and Technology of China 2Hangzhou Innovation Institute, Beihang University, Hangzhou, China |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described, nor does it contain explicit code release statements or repository links. |
| Open Datasets | Yes | We use two public datasets, 3D MNIST 1 and the aligned Model Net40 2 (Wu et al. 2015). 1https://www.kaggle.com/daavoo/3d-mnist/version/13 2http://modelnet.cs.princeton.edu/ |
| Dataset Splits | No | The paper specifies training and test set sizes for 3D MNIST (5000 and 1000) and ModelNet40 (9,743 and 2,468) but does not explicitly mention or detail a validation set split. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running its experiments. |
| Software Dependencies | No | The paper mentions various models and optimizers but does not specify any software dependencies with version numbers. |
| Experiment Setup | Yes | The two hyperparameters of λ1 and λ2 are 0.3 and 0.7, respectively. The values of the two parameters α and β are both 0.5. |