Model-aware privacy-preserving with start trigger method for person re-identification

Tongzhen Si, Penglei Li, Xiaohui Yang, Linkun Fan, Fazhi He

Abstract: Highlights•We are the first to study the effective model switch to control the system status.•We develop the universal adversarial algorithm to produce an ingenious start trigger.•We design a model-aware training strategy to perceive issued instructions.•We exhibit an in-depth study and prove MPST has the privacy protection ability.