Constructing Stronger and Faster Baselines for Skeleton-Based Action Recognition

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 45; no. 2; pp. 1474 - 1488
• Main Authors: Song, Yi-Fan, Zhang, Zhang, Shan, Caifeng, Wang, Liang
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.02.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Action recognition; Activity recognition; Compounds; Computational modeling; Convolution; Datasets; EfficientNet; Feature extraction; graph convolutional network; separable convolution; Skeleton; skeleton sequence; Source code; Task analysis; Training

One essential problem in skeleton-based action recognition is how to extract discriminative features over all skeleton joints. However, the complexity of the recent State-Of-The-Art (SOTA) models for this task tends to be exceedingly sophisticated and over-parameterized. The low efficiency in model...