An Edge-device Based Fast Fall Detection Using Spatio-temporal Optical Flow Model

Yuchao Yang; Hongwei Ren; Chenghao Li; Chenchen Ding; Hao Yu

10.1109/EMBC46164.2021.9629840

2021-10-30

2021 43th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

1557-170X

1558-4615

978-1-7281-1180-3

2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)

5067-5071

2021-11-01

online

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

["The elderly fall detection is one critical function in health of the elderly. A real-time fall detection for the elderly has been a significant healthcare issue. The traditional video analysis on cloud has large communication overhead. In this paper, a fast fall detection system based on the spatiotemporal optical flow model is proposed, which is further deeply compressed by a structured tensorization towards an implementation on edge devices. Firstly, an object extractor is built to extract motion objects from video clips. The spatiotemporal optical flow model is formed to estimate optical flow fields of motion objects. It can extract features from objects and their corresponding optical flow fields. Then these two features are fused to form new spatio-temporal features. Finally, the tensor-compressed model processes the fused features to determine fall detection, where the strongest optical field would indicate the fall. We conduct experiments with Multicam and URFD datasets.","Clinical relevance- It demonstrates that the proposed model achieves the accuracy of 9623% and 9937%, respectively. Besides, it attains the inference speed of 833 FPS and storage reduction of 2109x. Our work is further implemented on an AI acceleration core based edge device, and the runtime is reduced by 921x.This high performance system can be applied to the field of clinical monitoring in the future."]

Performance evaluation Runtime Image edge detection Senior citizens Streaming media Feature extraction Real-time systems

第一 ; 通讯

英语

CPCI-S ; EI

National Natural Science Foundation of China (NSFC)[62034007]

Engineering

Engineering, Biomedical ; Engineering, Electrical & Electronic

WOS:000760910505002

20220811670011

人工提交

Southern University of Science and Technology

Yuchao Yang,Hongwei Ren,Chenghao Li,et al. An Edge-device Based Fast Fall Detection Using Spatio-temporal Optical Flow Model[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2021:5067-5071.