Can I read Tennis Action Recognition Based on Multi-Branch Mixed Attention on EtoBox?

Tennis Action Recognition Based on Multi-Branch Mixed Attention by Xianwei Zhou; Weitao Chen; Zhenfeng Li; Yuan Li; Jiale Lei; Songsen Yu is a book available to read on EtoBox.

What is Tennis Action Recognition Based on Multi-Branch Mixed Attention about?

Tennis action recognition is a challenging problem due to its inherent characteristics such as high speed and large amplitude of actions. In this paper, an end-to-end, Multi-Branch Mixed Attentionbased network model (MBMA-Net) is proposed to tackle those challenges. Specifically, the model is designed to: (a) capture the interdependencies between channel features using channel attention; (b) improve the spatial receptive field to better filter spatial features by spatial attention;and (c) model the actions between consecutive frames using motion excitation to improve the accuracy of action feature extraction and recognition. The experimental results on dataset THETIS show that MBMA-Net achieves an accuracy of 0.8698, 5.35% higher than other baseline models for tennis action recognition and the model is further validated through various ablation experiments.

Author

Xianwei Zhou; Weitao Chen; Zhenfeng Li; Yuan Li; Jiale Lei; Songsen Yu

Publisher

Springer International Publishing

Published

2023

Language

EN

ISBN

9783031402852

Subjects

Computer Science, Business, Science