Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition

Ordonez Morales, Francisco Javier and Roggen, Daniel (2016) Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors, 16 (1). pp. 1-25. ISSN 1424-8220

[img] PDF - Published Version
Available under License Creative Commons Attribution.

Download (2MB)

Abstract

Human activity recognition (HAR) tasks have traditionally been solved using engineered features obtained by heuristic processes. Current research suggests that deep convolutional neural networks are suited to automate feature extraction from raw sensor inputs. However, human activities are made of complex sequences of motor movements, and capturing this temporal dynamics is fundamental for successful HAR. Based on the recent success of recurrent neural networks for time series domains, we propose a generic deep framework for activity recognition based on convolutional and LSTM recurrent units, which: (i) is suitable for multimodal wearable sensors; (ii) can perform sensor fusion naturally; (iii) does not require expert knowledge in designing features; and (iv) explicitly models the temporal dynamics of feature activations. We evaluate our framework on two datasets, one of which has been used in a public activity recognition challenge. Our results show that our framework outperforms competing deep non-recurrent networks on the challenge dataset by 4% on average; outperforming some of the previous reported results by up to 9%. Our results show that the framework can be applied to homogeneous sensor modalities, but can also fuse multimodal sensors to improve performance. We characterise key architectural hyperparameters’ influence on performance to provide insights about their optimisation.

Item Type: Article
Keywords: human activity recognition, wearable sensors, deep learning, machine learning, sensor fusion, LSTM, neural network
Schools and Departments: School of Engineering and Informatics > Engineering and Design
Subjects: T Technology > T Technology (General) > T0055.4 Industrial engineering. Management engineering > T0058.5 Information technology
T Technology > T Technology (General) > T0055.4 Industrial engineering. Management engineering > T0059.7 Human engineering in industry. Man-machine systems
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics > TK7885 Computer engineering. Computer hardware
Related URLs:
Depositing User: Daniel Roggen
Date Deposited: 18 Jan 2016 12:54
Last Modified: 06 Mar 2017 16:20
URI: http://sro.sussex.ac.uk/id/eprint/59271

View download statistics for this item

📧 Request an update
Project NameSussex Project NumberFunderFunder Ref
Is deep learning useful for wearable activity recognition?G1460GOOGLEUnset
Lifelearn: Unbounded activity and context awarenessG1786EPSRC-ENGINEERING & PHYSICAL SCIENCES RESEARCH COUNCILEP/N007816/1