DeepAI AI Chat
Log In Sign Up

A Perturbation Constrained Adversarial Attack for Evaluating the Robustness of Optical Flow

by   Jenny Schmalfuss, et al.
University of Stuttgart

Recent optical flow methods are almost exclusively judged in terms of accuracy, while analyzing their robustness is often neglected. Although adversarial attacks offer a useful tool to perform such an analysis, current attacks on optical flow methods rather focus on real-world attacking scenarios than on a worst case robustness assessment. Hence, in this work, we propose a novel adversarial attack - the Perturbation Constrained Flow Attack (PCFA) - that emphasizes destructivity over applicability as a real-world attack. More precisely, PCFA is a global attack that optimizes adversarial perturbations to shift the predicted flow towards a specified target flow, while keeping the L2 norm of the perturbation below a chosen bound. Our experiments not only demonstrate PCFA's applicability in white- and black-box settings, but also show that it finds stronger adversarial samples for optical flow than previous attacking frameworks. Moreover, based on these strong samples, we provide the first common ranking of optical flow methods in the literature considering both prediction quality and adversarial robustness, indicating that high quality methods are not necessarily robust. Our source code will be publicly available.


page 1

page 6

page 10

page 16

page 18

page 19


Consistent Semantic Attacks on Optical Flow

We present a novel approach for semantically targeted adversarial attack...

Attacking Motion Estimation with Adversarial Snow

Current adversarial attacks for motion estimation (optical flow) optimiz...

CosPGD: a unified white-box adversarial attack for pixel-wise prediction tasks

While neural networks allow highly accurate predictions in many tasks, t...

What Causes Optical Flow Networks to be Vulnerable to Physical Adversarial Attacks

Recent work demonstrated the lack of robustness of optical flow networks...

Attacking Optical Flow

Deep neural nets achieve state-of-the-art performance on the problem of ...

Investigating Robustness of Adversarial Samples Detection for Automatic Speaker Verification

Recently adversarial attacks on automatic speaker verification (ASV) sys...

Movement Tracking by Optical Flow Assisted Inertial Navigation

Robust and accurate six degree-of-freedom tracking on portable devices r...