AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

Efficient Decision-based Black-box Patch Attacks on Video Recognition

Kaixun Jiang, Zhaoyu Chen, Hao Huang, Jiafeng Wang, Dingkang Yang, Bo Li, Yan Wang, Wenqiang Zhang

ICCV 2023 · Adversarial robustness and security › Black-box and transfer attacks · approach: Robustness benchmarks and defenses · 32 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use evolutionary STDE instead of reinforcement-learning TPA in decision-based settingSTDE · UCF-101 untargeted (C3D)fooling rate: 56.40 -> 100Improved
use STDE instead of score-based BSCA while using only hard labelsSTDE · UCF-101 untargeted (C3D)fooling rate: 70.90 (BSCA*) -> 100Improved
port STDE to bullet-screen patches (STDE-W) versus BSCABSCA · UCF-101 untargeted (C3D)fooling rate / queries: 59.4 -> 65.3; AQN 4821 -> 1229Improved
fill patch texture with target video instead of white (ablation)STDE bullet screen · UCF-101 targeted (C3D)fooling rate: 17.8 -> 23.8Improved
add temporal difference keyframe selection (ablation)STDE · UCF-101 targeted (C3D)AOA: 24.90 -> 18.70Improved
add temporal difference keyframe selection (ablation)STDE · UCF-101 untargeted (C3D)AOA: 4.19 -> 4.06Improved
add crossover operation (ablation)STDE · UCF-101 targeted (C3D)AOA: AOA increases by 20% without crossoverImproved
use l0 norm instead of l2 norm in fitness function (ablation)STDE · UCF-101 (C3D)AOA: AOA drops by 1% targeted, 0.16% untargetedImproved
use target video texture instead of Gaussian noise or monochrome blocks (ablation)STDE · UCF-101 untargeted (C3D)AOA: 5% increase in AOA with alternativesImproved
choose different source/target videos (ablation)STDE · UCF-101 (C3D, NL, TPN)fooling rate: 100 on all modelsNo clear effect
apply patch defenses LGS and DW to inputSTDE · NL, UCF-101fooling rate: 100 -> 98 (DW untargeted)No clear effect
use STDE instead of BSCA for image qualityBSCA · UCF-101 (C3D)PSNR: 22.74 -> 34.04Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy