System and Method for Video Backdoor Attack

A method for video backdoor attack, including

• a trigger generation step for generating a universal adversarial trigger pattern specific to a task,
• an adversarial perturbation step for producing videos with manipulated features;
• and
• an poisoning and inference step for injecting the generated trigger into perturbed videos as poisoned samples for training;
• wherein the trigger pattern is patched and optimized on videos from all non-target classes but relabeled to a target class, and
• the trigger pattern is a universal adversarial trigger pattern generated by minimizing the cross-entropy loss.

A system for video backdoor attack, including

• a trigger generation module for generating a universal adversarial trigger pattern specific to a task,
• an adversarial perturbation module for producing videos with manipulated features;
• and
• a poisoning and inference module for injecting the generated trigger pattern into perturbed videos as poisoned samples for training.