Jung-Woo Chang (University of California San Diego), Mojan Javaheripi (University of California San Diego), Seira Hidano (KDDI Research, Inc.), Farinaz Koushanfar (University of California San Diego)

Video compression plays a crucial role in video streaming and classification systems by maximizing the end-user quality of experience (QoE) at a given bandwidth budget.

In this paper, we conduct the first systematic study for adversarial attacks on deep learning-based video compression and downstream classification systems. Our attack framework, dubbed RoVISQ, manipulates the Rate-Distortion (R-D) relationship of a video compression model to achieve one or both of the following goals: (1) increasing the network bandwidth, (2) degrading the video quality for end-users. We further devise new objectives for targeted and untargeted attacks to a downstream video classification service. Finally, we design an input-invariant perturbation that universally disrupts video compression and classification systems in real time. Unlike previously proposed attacks on video classification, our adversarial perturbations are the first to withstand compression.

We empirically show the resilience of RoVISQ attacks against various defenses, i.e., adversarial training, video denoising, and JPEG compression. Our extensive experimental results on various video datasets show RoVISQ attacks deteriorate peak signal-to-noise ratio by up to 5.6dB and the bit-rate by up to ~ 2.4 times while achieving over 90% attack success rate on a downstream classifier. Our user study further demonstrates the effect of RoVISQ attacks on users’ QoE. We provide several example attacked videos used in our survey on https://sites.google.com/view/demo-of-rovisq/home.

View More Papers

Hope of Delivery: Extracting User Locations From Mobile Instant...

Theodor Schnitzler (Research Center Trustworthy Data Science and Security, TU Dortmund, and Ruhr-Universität Bochum), Katharina Kohls (Radboud University), Evangelos Bitsikas (Northeastern University and New York University Abu Dhabi), Christina Pöpper (New York University Abu Dhabi)

Read More

Non-Interactive Privacy-Preserving Sybil-Free Authentication Scheme in VANETs

Mahdi Akil (Karlstad University), Leonardo Martucci (Karlstad University), Jaap-Henk Hoepman (Radboud University)

Read More

Semi-Automated Synthesis of Driving Rules

Diego Ortiz, Leilani Gilpin, Alvaro A. Cardenas (University of California, Santa Cruz)

Read More