Driver Action Recognition Based on Dynamic Adaptive Transformer

EasyChair Preprint 10777, version 1

Abstract

In industrial-grade applications, the efficiency of algorithms andmodels takes precedence, ensuring a certain level of performance while aligningwith the specific requirements of the application and the capabilities of the underlying equipment. In recent years, the Vision Transformer has been introduced as a powerful approach to significantly improve recognition accuracy invarious tasks. However, it faces challenges concerning portability, as well ashigh computational and input requirements. To tackle these issues, a dynamicadaptive transformer (DAT) has been proposed. This innovative method involves dynamic parameter pruning, enabling the trained Vision Transformer toadapt effectively to different tasks. Experimental results demonstrate that thedynamic adaptive transformer (DAT) is capable of reducing the model's parameters and Gmac with minimal accuracy loss.

Keyphrases: Driver action recognition, computer vision, deep learning, dynamic adaptive network, spatiotemporal attention

@booklet{EasyChair:10777,
  author    = {Junqi Li and Tao Peng and Junjie Huang and Junping Liu and Xinrong Hu and Zili Zhang and Yu Mao},
  title     = {Driver Action Recognition Based on Dynamic Adaptive Transformer},
  howpublished = {EasyChair Preprint 10777},
  year      = {EasyChair, 2023}}