Skip to main navigation Skip to search Skip to main content

Comparison of Knowledge Distillation and Binarized Neural Networks for Human Activity Recognition Using Radar Data

  • Samuel Kakuba
  • , Savina Jassica Colaco
  • , Jung Hwan Kim
  • , Young Jin Yoon
  • , Dong Seog Han
  • Kyungpook National University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Though deep learning models exhibit good performance, it’s usually challenging to deploy them in low-resource devices and embedded platforms for human-to-machine interaction that are often faced with challenges of limited computational resources. In this paper, we carried out comparative experiments to ascertain the robustness of binarized neural networks (BNNs) and deep learning models that use knowledge distillation in relation to their performance on the University of Glasgow radar data for activity recognition. We evaluated the performance of three ImageNet models and used the best deep learning model among them as a teacher to aid learning for a low-parameterized model (student) in a knowledge distillation paradigm. We also evaluated the performance of BNNs on the same dataset. While both approaches exhibit comparable performance, the student model size and loss values are far smaller than the BNN model. However, the BNN model exhibits a better performance in terms of accuracy and confusion ratio which makes the choice of these deep learning approaches for low-resource devices a trade-off between accuracy and model size.

Original languageEnglish
Title of host publicationAdvances in Intelligent Information Hiding and Multimedia Signal Processing, Volume 1 - Proceeding of the 19th International Conference on IIH-MSP in conjunction with 11th International Conference on Orange Technology, Applications and Tools
EditorsShih-Pang Tseng, Anand Paul, Jeng-Shyang Pan, Margarita Favorskaya
PublisherSpringer Science and Business Media Deutschland GmbH
Pages233-240
Number of pages8
ISBN (Print)9789819787630
DOIs
StatePublished - 2025
Event19th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2023, in conjunction with the 11th International Conference on Orange Technology, Applications, and Tools, ICOT 2023 - Daegu, Korea, Republic of
Duration: 5 Dec 20237 Dec 2023

Publication series

NameSmart Innovation, Systems and Technologies
Volume415
ISSN (Print)2190-3018
ISSN (Electronic)2190-3026

Conference

Conference19th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2023, in conjunction with the 11th International Conference on Orange Technology, Applications, and Tools, ICOT 2023
Country/TerritoryKorea, Republic of
CityDaegu
Period5/12/237/12/23

Keywords

  • Activity recognition
  • Binarized neural network
  • Knowledge distillation
  • Radar data

Fingerprint

Dive into the research topics of 'Comparison of Knowledge Distillation and Binarized Neural Networks for Human Activity Recognition Using Radar Data'. Together they form a unique fingerprint.

Cite this