Detecting and Segmenting Adversarial Graphics Patterns from Images
This addresses a practical security issue for social media platforms facing adversarial image uploads, though it appears incremental as it builds on existing segmentation algorithms.
The paper tackles the problem of detecting and segmenting artificial graphics patterns that hackers add to inappropriate images to fool automated screening systems in social media, proposing a new segmentation method that outperforms baselines with promising generalization capability.
Adversarial attacks pose a substantial threat to computer vision system security, but the social media industry constantly faces another form of "adversarial attack" in which the hackers attempt to upload inappropriate images and fool the automated screening systems by adding artificial graphics patterns. In this paper, we formulate the defense against such attacks as an artificial graphics pattern segmentation problem. We evaluate the efficacy of several segmentation algorithms and, based on observation of their performance, propose a new method tailored to this specific problem. Extensive experiments show that the proposed method outperforms the baselines and has a promising generalization capability, which is the most crucial aspect in segmenting artificial graphics patterns.