Skeleton-Robustness-Benchmark(RobustBenchHAR)

RobustBenchHAR is a pytorch framework to boost and evaluate the adversarial transferability for Human skeletal behavior recognition. Official code for the paper:

ICLR2025 TASAR: Transfer-based Attack on Skeletal Action Recognition

Key Features of RobustBenchHAR:

• A benchmark for evaluating existing transfer-based attacks in human Activity Recognition (HAR): RobustBenchHAR ensembles existing transfer-based attacks including several types and fairly evaluates various transfer-based attacks under the same setting.
• Evaluate the robustness of various models and datasets: RobustBenchHAR provides a plug-and-play interface to verify the robustness of models on different data sets.
• A summary of transfer-based attacks and defenses: RobustBenchHAR reviews numerous transfer-based attacks and adversarial defenses, making it easy to get the whole picture of transfer-based attacks for practitioners.

Usage

The complete code will be available soon.