Activity Recognition Based on FR-CNN and Attention-Based LSTM Network

Tan Hsu Tan, Ching Jung Huang, Munkhjargal Gochoo, Yung Fu Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

A human activity recognition (HAR) based on the Faster Region-based Convolutional Neural Network (FR-CNN) and attention-based LSTM networks is proposed in this paper. A new structure of posture vector is generated by extracting skeleton joints of human movement using the pre-trained FR-CNN model. The Cornell Activity Dataset (CAD-60) is employed in the training and test phases. An attention-based bidirectional LSTM (Bi-LSTM) network is presented for activity classification. Experimental result shows that the attention-based Bi-LSTM network achieves the precision and recall rate of 97.02% and 96.83%, respectively, in recognizing twelve activities. The result is superior to the other existing systems, indicating the application potential of our work.

Original languageEnglish
Title of host publication2021 30th Wireless and Optical Communications Conference, WOCC 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages146-149
Number of pages4
ISBN (Electronic)9781665427722
DOIs
Publication statusPublished - 2021
Event30th Wireless and Optical Communications Conference, WOCC 2021 - Taipei, Taiwan, Province of China
Duration: Oct 7 2021Oct 8 2021

Publication series

Name2021 30th Wireless and Optical Communications Conference, WOCC 2021

Conference

Conference30th Wireless and Optical Communications Conference, WOCC 2021
Country/TerritoryTaiwan, Province of China
CityTaipei
Period10/7/2110/8/21

Keywords

  • attention-based bidirectional LSTM
  • CAD-60
  • deep learning
  • FR-CNN
  • human activity recognition

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Computer Networks and Communications
  • Signal Processing

Fingerprint

Dive into the research topics of 'Activity Recognition Based on FR-CNN and Attention-Based LSTM Network'. Together they form a unique fingerprint.

Cite this