Ensembled Transfer Learning Based Multichannel Attention Networks for Human Activity Recognition in Still Images

Human activity recognition is one of the most difficult tasks in computer vision. Due to the lack of time information, detecting human activities from still photos is more difficult than sensor-based or video-based techniques. Recently, various deep learning based solutions are being proposed one af...

Full description

Bibliographic Details

Published in:	IEEE Access
Main Authors:	Koki Hirooka, Md. Al Mehedi Hasan, Jungpil Shin, Azmain Yakin Srizon
Format:	Article
Language:	English
Published:	IEEE 2022-01-01
Subjects:	Human activity recognition multi-channel attention module ensembling InceptionV3 Xception InceptionResnetV2
Online Access:	https://ieeexplore.ieee.org/document/9765465/