A Bayesian Network for the Classification of Human Motion as Observed by Distributed Radar

Svenningsson, Peter; Fioranelli, Francesco; Yarovoy, Alexander; Martone, Anthony F.

doi:10.1109/taes.2022.3177589

Cited by 6 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, temporal resolution changes are achieved by downsampling along the temporal dimension using ratios of [0.5, 0.25, 0.125]. Doppler resolution changes are achieved by changing the number of FFT points from 128 to [64,32,16]. Drop ratios are selected from [10%, 20%, 30%, 40%, 50%].…”

Section: Evaluation Methods 1) Classification Accuracymentioning

confidence: 99%

Radar-Based Swimming Activity Recognition with Temporal Dynamic Convolution and Spectral Data Augmentation

Zhou,

Yu,

Lopez-Benitez

et al. 2024

Preprint

View full text Add to dashboard Cite

Radar-based human activity recognition (HAR) is a popular area of research. In this paper, we investigate methods to improve the generalization of micro-Doppler-based swimming activity recognition. We identify three main challenges to this task: a small dataset lacking motion diversity, inaccurate period estimation, and inefficient network design that does not take into account the unique characteristics of spectrograms. To address the limited motion diversity, we propose a spectral data augmentation tailored for micro-Doppler spectrograms, including positive augmentations that account for physical fidelity and negative augmentations that penalize the unrealistic examples. We also investigate self-supervised pre-training to effectively use these negative augmentations. To address inaccurate period estimation, we introduce a segmentation approach based on energy distribution to handle temporal period variation. To exploit the spreading pattern of limb motion in the Doppler dimension and the continuous properties of torso motion in the temporal dimension, we design a module consisting of both 2D convolution and 1D temporal dynamic convolution to serve as a feature extractor. Our evaluation on a self-collected swimming activity recognition dataset shows that our model achieves the best classification accuracy and robustness to corruptions, even compared to much larger models and multi-domain fusion models.

show abstract

Section: Evaluation Methods 1) Classification Accuracymentioning

confidence: 99%

Radar-Based Swimming Activity Recognition with Temporal Dynamic Convolution and Spectral Data Augmentation

Zhou,

Yu,

Lopez-Benitez

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Svenningsson et al [55] proposed another processing approach for the TU Delft data set in 2022. In this paper, a Bayesian network is proposed, where in a recursive filtering algorithm the target's state (position, velocity, heading and turn rate) and motion class were jointly estimated.…”

Section: The History Of Continuous Human Activity Recognitionmentioning

confidence: 99%

“…This is accomplished e.g. by means of a human ethogram [44], [45] or Markov chain [55] and is based on basic assumptions about human kinematics. Finally, a notable trend in the ensemble of works under review here is the strong emphasis on automated feature construction.…”

Section: The History Of Continuous Human Activity Recognitionmentioning

confidence: 99%

A Survey on Radar-Based Continuous Human Activity Recognition

Ullmann¹,

Guendel

Kruse

et al. 2023

IEEE J. Microw.

View full text Add to dashboard Cite

Radar-based human motion and activity recognition is currently a topic of great research interest, as the aging population increases and older individuals prefer an independent lifestyle. This technology has a wide range of applications, such as fall detection in assisted living, gesture recognition for human-machine interfaces, and many more. Numerous studies exist on various approaches for radar-based activity capture and classification. However, most of these employ rather artificial data, often obtained in laboratory environments, and typically collected under particular conditions. Specifically, most research so far has aimed at distinguishing a predefined set of single activities with a defined start, stop and duration. This paper aims at drawing the attention to a so far less researched issue, one that will be of vital importance for future real-world application of radar-based human activity recognition: continuous activity recognition, i.e. recognizing specific activities in a stream of several sequential activities with unknown duration and arbitrary transitions between different classes of activities. A review on the current state of the art in this relatively new topic is given, followed by a discussion on future research directions.INDEX TERMS Radar applications, radar signal processing, continuous human activity recognition, activities of daily living.This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

show abstract

RMPCT-Net: a multi-channel parallel CNN and transformer network model applied to HAR using FMCW radar

Gu,

Chen,

Chen

et al. 2023

SIViP

View full text Add to dashboard Cite

A Bayesian Network for the Classification of Human Motion as Observed by Distributed Radar

Cited by 6 publications

References 32 publications

Radar-Based Swimming Activity Recognition with Temporal Dynamic Convolution and Spectral Data Augmentation

Radar-Based Swimming Activity Recognition with Temporal Dynamic Convolution and Spectral Data Augmentation

A Survey on Radar-Based Continuous Human Activity Recognition

RMPCT-Net: a multi-channel parallel CNN and transformer network model applied to HAR using FMCW radar

Contact Info

Product

Resources

About