Fooling Polarization-Based Vision Using Locally Controllable Polarizing Projection
Zhuoxiao Li, Zhihang Zhong, Shohei Nobuhara, Ko Nishino, Yinqiang Zheng
Abstract
Polarization is a fundamental property of light that encodes abundant information regarding surface shape, material, illumination and viewing geometry. The computer vision community has witnessed a blossom of polarizationbased vision applications, such as reflection removal, shape-from-polarization (SfP), transparent object segmentation and color constancy, partially due to the emergence of single-chip mono/color polarization sensors that make polarization data acquisition easier than ever. However, is polarization-based vision vulnerable to adversarial attacks? If so, is that possible to realize these adversarial attacks in the physical world, without being perceived by human eyes? In this paper, we warn the community of the vulnerability of polarization-based vision, which can be more serious than RGB-based vision. By adapting a commercial LCD projector, we achieve locally controllable polarizing projection, which is successfully utilized to fool state-of-theart polarization-based vision algorithms for glass segmentation and SfP. Compared with existing physical attacks on RGB-based vision, which always suffer from the trade-off between attack efficacy and eye conceivability, the adversarial attackers based on polarizing projection are contactfree and visually imperceptible, since naked human eyes can rarely perceive the difference of viciously manipulated polarizing light and ordinary illumination. This poses unprecedented risks on polarization-based vision, for which due attentions should be paid and counter measures be considered.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Builds on12
- Towards Evaluating the Robustness of Neural NetworksNicholas Carlini, David A. WagnerS&P 2017 · 9,786 citations
- Accessorize to a Crime: Real and Stealthy Attacks on State-of-the-Art Face RecognitionMahmood Sharif, Sruti Bhagavatula, Lujo Bauer, Michael K. ReiterCCS 2016 · 1,765 citations
- Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural PhenomenonYiqi Zhong, Xianming Liu, Deming Zhai, Junjun Jiang et al.CVPR 2022 · 148 citations
- Glass Segmentation using Intensity and Spectral Polarization CuesHaiyang Mei, Bo Dong, Wen Dong, Jiaxi Yang et al.CVPR 2022 · 91 citations
- Shape from Polarization for Complex Scenes in the WildChenyang Lei, Chenyang Qi, Jiaxin Xie, Na Fan et al.CVPR 2022 · 60 citations
Related papers
- Degree-of-linear-polarization-based Color ConstancyTaishi Ono, Yuhi Kondo, Legong Sun, Teppei Kurita et al.CVPR 2022 · 23 citations
- SPIDeRS: Structured Polarization for Invisible Depth and Reflectance SensingTomoki Ichikawa, Shohei Nobuhara, Ko NishinoCVPR 2024
- Polarization State Tracing for Reflection Removal and Color-Consistent ReconstructionDongyue Wang, Yang Lu, Jiandong TianCVPR 2026
- Deep Polarization Cues for Transparent Object SegmentationAgastya Kalra, Vage Taamazyan, Supreeth Krishna Rao, Kartik Venkataraman et al.CVPR 2020
- PolarDepth: Monocular Transparent Object Depth from Polar-Physics PriorsWen Dong, Haiyang Mei, Yinglian Ji, Zijun Zhang et al.ICML 2026
