Zhipeng Wei
Jilin University
19 Papers
4 Citations
Zhipeng Wei is an academic researcher from Jilin University. The author has contributed to research in topics: Computer science & Adversarial system. The author has an hindex of 4, co-authored 8 publications. Previous affiliations of Zhipeng Wei include Fudan University.
Chat about Author
Papers
Heuristic Black-Box Adversarial Attacks on Video Recognition Models
Zhipeng Wei,Jingjing Chen,Xingxing Wei,Linxi Jiang,Tat-Seng Chua,Fengfeng Zhou,Yu-Gang Jiang +6 more
- 03 Apr 2020
TL;DR: A heuristic black-box adversarial attack model that generates adversarial perturbations only on the selected frames and regions is proposed that can significantly reduce the computation cost and lead to more than 28% reduction in query numbers for the untargeted attack on both datasets.
Zero-Shot Ingredient Recognition by Multi-Relational Graph Convolutional Network
Jingjing Chen,Liangming Pan,Zhipeng Wei,Xiang Wang,Chong-Wah Ngo,Tat-Seng Chua +5 more
- 03 Apr 2020
TL;DR: Multi-relational GCN (graph convolutional network) is introduced that integrates ingredient hierarchy, attribute as well as co-occurrence for zero-shot ingredient recognition and sheds light on zero- shot ingredients recognition.
•Posted Content
Towards Transferable Adversarial Attacks on Vision Transformers
TL;DR: Wang et al. as mentioned in this paper proposed a dual attack framework, which contains a Pay No Attention (PNA) attack and a PatchOut attack, to improve the transferability of adversarial samples across different vision transformers.
45
•Posted Content
Boosting the Transferability of Video Adversarial Examples via Temporal Translation
TL;DR: Wang et al. as mentioned in this paper proposed a temporal translation attack method to boost the transferability of video adversarial examples for black-box attacks on video recognition models, which achieved a 61.56% average attack success rate on the Kinetics-400 and 48.60% on the UCF-101 datasets.
18
Enhancing the Self-Universality for Transferable Targeted Attacks
Zhipeng Wei,Jingjing Chen,Zuxuan Wu,Yueping Jiang +3 more
- 08 Sep 2022
TL;DR: Li et al. as discussed by the authors proposed a self-universality (SU) attack to make the perturbation to be agnostic to different local regions within one image, which is called as self universality, and introduced a feature similarity loss that encourages the learned perturbations to be universal by maximizing the feature similarity between adversarial perturbed global images and randomly cropped local regions.
15