SkeletonVis: Interactive Visualization for Understanding Adversarial Attacks on Human Action Recognition Models
Not on the map. Why?
Abstract (arXiv)
Skeleton-based human action recognition technologies are increasingly used in video based applications, such as home robotics, healthcare on aging population, and surveillance. However, such models are vulnerable to adversarial attacks, raising serious concerns for their use in safety-critical applications. To develop an effective defense against attacks, it is essential to understand how such attacks mislead the pose detection models into making incorrect predictions. We present SkeletonVis, the first interactive system that visualizes how the attacks work on the models to enhance human understanding of attacks.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 7 of its 11 references that are in the map. First posted on arXiv 26 Jan 2021.
Builds on
7 of 11 references.
References · 7
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
- 2019Adversarial Attack on Skeleton-based Human Action Recognitioncited by 32
- 2020Bluff: Interactively Deciphering Adversarial Attacks on Deep Neural Networkscited by 12
- 2020UnMask: Adversarial Detection and Defense Through Robust Feature Alignmentcited by 8
- 2020Massif: Interactive Interpretation of Adversarial Attacks on Deep Learningcited by 4
- 2019NeuralDivergence: Exploring and Understanding Neural Networks by Comparing Activation Distributionscited by 1
Built on by
Citing papers per year: 2023: 1.
1 paper.