LORTSAR: Low-Rank Transformer for Skeleton-based Action Recognition

TL;DR

LORTSAR employs SVD-based low-rank approximation and fine-tuning to significantly reduce Transformer model sizes for skeleton-based action recognition, maintaining or improving accuracy on benchmark datasets.

Contribution

This paper introduces LORTSAR, a novel low-rank Transformer compression method that effectively reduces model size while preserving or enhancing recognition performance.

Findings

Substantial parameter reduction with negligible accuracy loss.

Performance improvement on NTU RGB+D datasets.

Enhanced model efficiency and sustainability.

Abstract

The complexity of state-of-the-art Transformer-based models for skeleton-based action recognition poses significant challenges in terms of computational efficiency and resource utilization. In this paper, we explore the application of Singular Value Decomposition (SVD) to effectively reduce the model sizes of these pre-trained models, aiming to minimize their resource consumption while preserving accuracy. Our method, LORTSAR (LOw-Rank Transformer for Skeleton-based Action Recognition), also includes a fine-tuning step to compensate for any potential accuracy degradation caused by model compression, and is applied to two leading Transformer-based models, "Hyperformer" and "STEP-CATFormer". Experimental results on the "NTU RGB+D" and "NTU RGB+D 120" datasets show that our method can reduce the number of model parameters substantially with negligible degradation or even performance increase in recognition accuracy. This confirms that SVD combined with post-compression fine-tuning can boost model efficiency, paving the way for more sustainable, lightweight, and high-performance technologies in human action recognition.