Detecting Adversarial Attacks On Audiovisual Speech Recognition
This addresses a security threat for multi-modal deep learning systems, but it is incremental as it applies a known detection concept to a new domain.
The paper tackled the problem of detecting adversarial attacks on audiovisual speech recognition models by proposing a method based on temporal correlation between audio and video streams, showing it is effective in experiments on GRID and LRW datasets.
Adversarial attacks pose a threat to deep learning models. However, research on adversarial detection methods, especially in the multi-modal domain, is very limited. In this work, we propose an efficient and straightforward detection method based on the temporal correlation between audio and video streams. The main idea is that the correlation between audio and video in adversarial examples will be lower than benign examples due to added adversarial noise. We use the synchronisation confidence score as a proxy for audiovisual correlation and based on it we can detect adversarial attacks. To the best of our knowledge, this is the first work on detection of adversarial attacks on audiovisual speech recognition models. We apply recent adversarial attacks on two audiovisual speech recognition models trained on the GRID and LRW datasets. The experimental results demonstrate that the proposed approach is an effective way for detecting such attacks.