# Left-right Discrepancy for Adversarial Attack on Stereo Networks

arXiv: https://arxiv.org/abs/2401.07188 · HTML: https://mlsanity.com/paper/2401.07188/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 35 references that are in the map.
- First posted on arXiv 14 Jan 2024.

## Builds on (most cited first, 29 of 29)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [DeepFool: a simple and accurate method to fool deep neural networks](https://mlsanity.com/paper/1511.04599/) (2015)
- [Adversarial Machine Learning at Scale](https://mlsanity.com/paper/1611.01236/) (2016)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [GhostNet: More Features from Cheap Operations](https://mlsanity.com/paper/1911.11907/) (2019)
- [Adversarial Examples for Semantic Segmentation and Object Detection](https://mlsanity.com/paper/1703.08603/) (2017)
- [AANet: Adaptive Aggregation Network for Efficient Stereo Matching](https://mlsanity.com/paper/2004.09548/) (2020)
- [Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlation](https://mlsanity.com/paper/2203.11483/) (2022)
- [Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution](https://mlsanity.com/paper/1904.05049/) (2019)
- [Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networks](https://mlsanity.com/paper/2303.03667/) (2023)
- [On the Robustness of Semantic Segmentation Models to Adversarial Attacks](https://mlsanity.com/paper/1711.09856/) (2017)
- [Improving Black-box Adversarial Attacks with a Transfer-based Prior](https://mlsanity.com/paper/1906.06919/) (2019)
- [DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch](https://mlsanity.com/paper/1909.05845/) (2019)
- [Cross-Domain Transferability of Adversarial Perturbations](https://mlsanity.com/paper/1905.11736/) (2019)
- [Computing the Stereo Matching Cost with a Convolutional Neural Network](https://mlsanity.com/paper/1409.4326/) (2014)
- [PCW-Net: Pyramid Combination and Warping Cost Volume for Stereo Matching](https://mlsanity.com/paper/2006.12797/) (2020)
- [Physical Attack on Monocular Depth Estimation with Optimal Adversarial Patches](https://mlsanity.com/paper/2207.04718/) (2022)
- [Evaluating the Robustness of Semantic Segmentation for Autonomous Driving against Real-World Adversarial Patch Attacks](https://mlsanity.com/paper/2108.06179/) (2021)
- [Attacking Optical Flow](https://mlsanity.com/paper/1910.10053/) (2019)
- [Targeted Adversarial Perturbations for Monocular Depth Prediction](https://mlsanity.com/paper/2006.08602/) (2020)
- [Adversarial Attacks on Monocular Depth Estimation](https://mlsanity.com/paper/2003.10315/) (2020)
- [Stereopagnosia: Fooling Stereo Networks with Adversarial Perturbations](https://mlsanity.com/paper/2009.10142/) (2020)
- [Stereoscopic Universal Perturbations across Different Architectures and Datasets](https://mlsanity.com/paper/2112.06116/) (2021)
- [A Perturbation-Constrained Adversarial Attack for Evaluating the Robustness of Optical Flow](https://mlsanity.com/paper/2203.13214/) (2022)
- [Towards Understanding Adversarial Robustness of Optical Flow Networks](https://mlsanity.com/paper/2103.16255/) (2021)
- [Left-Right Comparative Recurrent Model for Stereo Matching](https://mlsanity.com/paper/1804.00796/) (2018)

## Built on by (most cited first, 2 of 2)

- [RobustSpring: Benchmarking Robustness to Image Corruptions for Optical Flow, Scene Flow and Stereo](https://mlsanity.com/paper/2505.09368/) (2025)
- [Pixel-Optimization-Free Patch Attack on Stereo Depth Estimation](https://mlsanity.com/paper/2506.17632/) (2025)

## Abstract (arXiv)

Stereo matching neural networks often involve a Siamese structure to extract intermediate features from left and right images. The similarity between these intermediate left-right features significantly impacts the accuracy of disparity estimation. In this paper, we introduce a novel adversarial attack approach that generates perturbation noise specifically designed to maximize the discrepancy between left and right image features. Extensive experiments demonstrate the superior capability of our method to induce larger prediction errors in stereo neural networks, e.g. outperforming existing state-of-the-art attack methods by 219% MAE on the KITTI dataset and 85% MAE on the Scene Flow dataset. Additionally, we extend our approach to include a proxy network black-box attack method, eliminating the need for access to stereo neural network. This method leverages an arbitrary network from a different vision task as a proxy to generate adversarial noise, effectively causing the stereo network to produce erroneous predictions. Our findings highlight a notable sensitivity of stereo networks to discrepancies in shallow layer features, offering valuable insights that could guide future research in enhancing the robustness of stereo vision systems.