Meta Generative Attack on Person Reidentification
This addresses security vulnerabilities in person re-identification systems, but it is incremental as it builds on existing adversarial attack methods.
The paper tackles the problem of adversarial attacks in person re-identification, specifically improving transferability across different models and datasets, and reports favorable results on benchmarks like Market-1501, DukeMTMC-reID, and MSMT-17.
Adversarial attacks have been recently investigated in person re-identification. These attacks perform well under cross dataset or cross model setting. However, the challenges present in cross-dataset cross-model scenario does not allow these models to achieve similar accuracy. To this end, we propose our method with the goal of achieving better transferability against different models and across datasets. We generate a mask to obtain better performance across models and use meta learning to boost the generalizability in the challenging cross-dataset cross-model setting. Experiments on Market-1501, DukeMTMC-reID and MSMT-17 demonstrate favorable results compared to other attacks.