skip to main content
10.1145/3376897.3377859acmconferencesArticle/Chapter ViewAbstractPublication PageshotmobileConference Proceedingsconference-collections
research-article

Improving Resource Efficiency of Deep Activity Recognition via Redundancy Reduction

Published: 03 March 2020 Publication History

Abstract

Compression methods for deep learning have been recently used to port deep neural networks into resource-constrained devices - such as digital gloves and smartwatches - for human activity recognition (HAR). While the results have been in favor of utilizing compressed models, we envision that the current paradigm of long and fixed-size overlapping sliding windows that permeate the literature of HAR contributes negatively toward the goal of more resource-efficient systems, as it induces redundancies in memory and computation. In this work, we provide a different perspective by demonstrating that memory footprint, computational expense, and possibly energy consumption can be dramatically spared by modifying the architecture of the neural networks and their training. It is achieved by enabling non-overlapping short sliding windows and skipping fine-grained features in favor of rough ones on certain occasions, thus reducing the demand for more powerful hardware. Compared with the state-of-the-art, our method is able to achieve comparable performance far more efficiently in terms of resource use.

References

[1]
C. Alippi, S. Disabato, and M. Roveri. 2018. Moving Convolutional Neural Networks to Embedded Systems: The Alexnet and VGG-16 Case. In IPSN '18. IEEE Press, Piscataway, NJ, USA, 212--223.
[2]
M. Alizadeh and N. D. Lane. 2018. Using Pre-trained Full-Precision Models to Speed Up Training Binary Networks For Mobile Devices. In MobiSys '18. ACM, New York, NY, USA, 528--528.
[3]
J. Chauhan, J. Rajasegaran, S. Seneviratne, A. Misra, A. Seneviratne, and Y. Lee. 2018. Performance Characterization of Deep Learning Models for Breathing-based Authentication on Resource-Constrained Devices. 2018 ACM IMWUT, Vol. 2, 4, Article 158 (Dec. 2018), bibinfonumpages24 pages.
[4]
R. Chavarriaga, H. Sagha, A. Calatroni, S. T. Digumarti, G. Tröster, J. del R. Millán, and D. Roggen. 2013. The Opportunity challenge: A benchmark database for on-body sensor-based activity recognition. Pattern Recognition Letters, Vol. 34, 15 (2013), 2033 -- 2042.
[5]
M. Edel and E. Köppe. 2016. Binarized-BLSTM-RNN based Human Activity Recognition. In IPIN'16. 1--7.
[6]
B. Fang, X. Zeng, and M. Zhang. 2018. NestDNN: Resource-Aware Multi-Tenant On-Device Deep Learning for Continuous Mobile Vision. In MobiCom '18. ACM, 115--127.
[7]
Y. Guan and T. Plötz. 2017. Ensembles of Deep LSTM Learners for Activity Recognition Using Wearables. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 1, 2, Article 11 (June 2017), bibinfonumpages28 pages.
[8]
N. D. Lane, S. Bhattacharya, A. Mathur, C. Forlivesi, and F. Kawsar. 2016. DXTK: Enabling Resource-efficient Deep Learning on Mobile and Embedded Devices with the DeepX Toolkit. In MobiCASE'16. 98--107.
[9]
J. Long, W. Sun, Z. Yang, O. I. Raymond, and B. Li. 2019. Dual Residual Network for Accurate Human Activity Recognition. CoRR, Vol. abs/1903.05359 (2019). arxiv: 1903.05359
[10]
F. J. Ordóñez and D. Roggen. 2016. Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition. Sensors, Vol. 16, 1 (2016).
[11]
D. Ravi, C. Wong, B. Lo, and G. Yang. 2016. Deep learning for human activity recognition: A resource efficient implementation on low-power devices. In 2016 IEEE BSN. 71--76.
[12]
A. Reiss and D. Stricker. 2012. Introducing a New Benchmarked Dataset for Activity Monitoring. In ISWC'16 . 108--109.
[13]
C. Xu, D. Chai, J. He, X. Zhang, and S. Duan. 2019. InnoHAR: A Deep Neural Network for Complex Human Activity Recognition. IEEE Access, Vol. 7 (2019), 9893--9902.
[14]
S. Yao, Y. Zhao, H. Shao, S. Liu, D. Liu, L. Su, and T. Abdelzaher. 2018. FastDeepIoT: Towards Understanding and Optimizing Neural Network Execution Time on Mobile and Embedded Devices. In Sensys '18. ACM, New York, NY, USA, 278--291.

Cited By

View all
  • (2023)On-Device Deep Learning for Mobile and Wearable Sensing Applications: A ReviewIEEE Sensors Journal10.1109/JSEN.2023.324085423:6(5501-5512)Online publication date: 15-Mar-2023
  • (2023)Building Lightweight Deep learning Models with TensorFlow Lite for Human Activity Recognition on Mobile DevicesAnnals of Telecommunications10.1007/s12243-023-00962-x78:11-12(687-702)Online publication date: 15-Jul-2023
  • (2020)Clownfish: Edge and Cloud Symbiosis for Video Stream Analytics2020 IEEE/ACM Symposium on Edge Computing (SEC)10.1109/SEC50012.2020.00012(55-69)Online publication date: Nov-2020
  1. Improving Resource Efficiency of Deep Activity Recognition via Redundancy Reduction

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM Conferences
    HotMobile '20: Proceedings of the 21st International Workshop on Mobile Computing Systems and Applications
    March 2020
    116 pages
    ISBN:9781450371162
    DOI:10.1145/3376897
    Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

    Sponsors

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 03 March 2020

    Permissions

    Request permissions for this article.

    Check for updates

    Author Tags

    1. deep learning
    2. human activity recognition
    3. resource-constrained devices

    Qualifiers

    • Research-article

    Funding Sources

    • European Union's Horizon 2020 Research and Innovation Programme
    • Business Finland

    Conference

    HotMobile '20
    Sponsor:

    Acceptance Rates

    Overall Acceptance Rate 96 of 345 submissions, 28%

    Upcoming Conference

    HOTMOBILE '25

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • Downloads (Last 12 months)7
    • Downloads (Last 6 weeks)0
    Reflects downloads up to 27 Jan 2025

    Other Metrics

    Citations

    Cited By

    View all
    • (2023)On-Device Deep Learning for Mobile and Wearable Sensing Applications: A ReviewIEEE Sensors Journal10.1109/JSEN.2023.324085423:6(5501-5512)Online publication date: 15-Mar-2023
    • (2023)Building Lightweight Deep learning Models with TensorFlow Lite for Human Activity Recognition on Mobile DevicesAnnals of Telecommunications10.1007/s12243-023-00962-x78:11-12(687-702)Online publication date: 15-Jul-2023
    • (2020)Clownfish: Edge and Cloud Symbiosis for Video Stream Analytics2020 IEEE/ACM Symposium on Edge Computing (SEC)10.1109/SEC50012.2020.00012(55-69)Online publication date: Nov-2020

    View Options

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    EPUB

    View this article in ePub.

    ePub

    Figures

    Tables

    Media

    Share

    Share

    Share this Publication link

    Share on social media