Adversarial Attack on Video Classification
2 papers with code • 0 benchmarks • 0 datasets
Use optimizer to add disturbe on video frame to fool the video classification systems. The key issue exist in tremendos calculation and the selection of key frame and key area.
Benchmarks
These leaderboards are used to track progress in Adversarial Attack on Video Classification
Most implemented papers
Attacking Video Recognition Models with Bullet-Screen Comments
On both UCF-101 and HMDB-51 datasets, our BSC attack method can achieve about 90\% fooling rate when attacking three mainstream video recognition models, while only occluding \textless 8\% areas in the video.
Prior-Guided Adversarial Initialization for Fast Adversarial Training
Based on the observation, we propose a prior-guided FGSM initialization method to avoid overfitting after investigating several initialization strategies, improving the quality of the AEs during the whole training process.
