From Movements to Metrics: Evaluating Explainable AI Methods in Skeleton-Based Human Activity Recognition
Author:
Pellano Kimji N.1ORCID, Strümke Inga2ORCID, Ihlen Espen A. F.1ORCID
Affiliation:
1. Department of Neuromedicine and Movement Science, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology, 7034 Trondheim, Norway 2. Department of Computer Science, Faculty of Information Technology and Electrical Engineering, Norwegian University of Science and Technology, 7034 Trondheim, Norway
Abstract
The advancement of deep learning in human activity recognition (HAR) using 3D skeleton data is critical for applications in healthcare, security, sports, and human–computer interaction. This paper tackles a well-known gap in the field, which is the lack of testing in the applicability and reliability of XAI evaluation metrics in the skeleton-based HAR domain. We have tested established XAI metrics, namely faithfulness and stability on Class Activation Mapping (CAM) and Gradient-weighted Class Activation Mapping (Grad-CAM) to address this problem. This study introduces a perturbation method that produces variations within the error tolerance of motion sensor tracking, ensuring the resultant skeletal data points remain within the plausible output range of human movement as captured by the tracking device. We used the NTU RGB+D 60 dataset and the EfficientGCN architecture for HAR model training and testing. The evaluation involved systematically perturbing the 3D skeleton data by applying controlled displacements at different magnitudes to assess the impact on XAI metric performance across multiple action classes. Our findings reveal that faithfulness may not consistently serve as a reliable metric across all classes for the EfficientGCN model, indicating its limited applicability in certain contexts. In contrast, stability proves to be a more robust metric, showing dependability across different perturbation magnitudes. Additionally, CAM and Grad-CAM yielded almost identical explanations, leading to closely similar metric outcomes. This suggests a need for the exploration of additional metrics and the application of more diverse XAI methods to broaden the understanding and effectiveness of XAI in skeleton-based HAR.
Funder
Research Council of Norway European Union
Reference25 articles.
1. Nguyen, T.N., Huynh, H.H., and Meunier, J. (2016). Skeleton-based abnormal gait detection. Sensors, 16. 2. Liu, C., Fu, R., Li, Y., Gao, Y., Shi, L., and Li, W. (2021). A self-attention augmented graph convolutional clustering networks for skeleton-based video anomaly behavior detection. Appl. Sci., 12. 3. Guo, J., Liu, H., Li, X., Xu, D., and Zhang, Y. (2021). An attention enhanced spatial–temporal graph convolutional LSTM network for action recognition in Karate. Appl. Sci., 11. 4. Skeleton-based motion prediction: A survey;Usman;Front. Comput. Neurosci.,2022 5. Commission, E. (2021). Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union legIslative Acts, Proposal, European Commission.
|
|