2D progressive fusion module for action recognition*

Zhongwei Shen

doi:10.1016/j.imavis.2021.104122

2D progressive fusion module for action recognition*

2021-05-01

5

PRE

AI

DOI：10.1016/j.imavis.2021.104122

原文链接

原文求助

摘要

En 中文

Network convergence as well as recognition accuracy are essential issues when applying Convolutional Neural Networks (CNN) to human action recognition. Most deep learning methods neglect model convergence when striving to improve the abstraction capability, thus degrading the performances sharply when computing resources are limited. To mitigate this problem, we propose a structure named 2D Progressive Fusion (2DPF) Module which is inserted after the 2D backbone CNN layers. 2DPF fuses features through a novel 2D convolution on the spatial and temporal dimensions called variation attenuating convolution and applies fusion techniques to improve the recognition accuracy and the convergency. Our experiments performed on several benchmarks (e.g., Something-Something V1&V2, Kinetics400 & 600, AViD, UCF101) demonstrate the effectiveness of the proposed method. ARTICLE INFO. (c) 2021 Published by Elsevier B.V.

Keyword:

Convergence

spatiotemporal modeling

2D CNN

action recognition

2D progressive fusion module for action recognition*

摘要

AI总结

期刊

机构

引用论文