Implementation study of wearable sensors for activity recognition systems

Rezaie, Hamed; Ghassemian, Hassan

doi:10.1049/htl.2015.0017

Cited by 12 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to both simple and recognition tests, in the choice tests, HRT is estimated from multiple stimuli and it is expected the reacting individual makes a choice from among multiple responses. Several systems were developed and reported to estimate the reaction time of drivers using virtual reality time [8], computer-based measurement of cognitive functioning [9] and wearable systems [10][11][12]. The latter ones usually incorporate motion sensors mounted on the reacting individual.…”

mentioning

confidence: 99%

Technique to estimate human reaction time based on visual perception

Abbasi-Kesbi

Memarzadeh-Tehran

Deen

2017

Healthcare Technology Letters

View full text Add to dashboard Cite

The design and implementation of a wearable system to estimate the human reaction time (HRT) to visual stimulus based on two identical wireless motion sensors are described. Each sensor incorporates a motion sensor (gyroscope), a processor and a transceiver operating at the industrial, scientific and medical frequency of 2.45 GHz. Relevant tests to estimate the HRT are performed in two different scenarios including simple and recognition tests for 90 pairs of measurements. The obtained results are compared with a computer-based system to determine the accuracy of the proposed system. The root mean square error, standard deviation error and mean error of the results are 2.88, 6.17 and 0.3 ms for simple test while for recognition test as low as 3.34, 7.83 and 0.35 ms, respectively. The outcomes of the HRT estimation tests confirm HRT can increase by 40–87% due to increased fatigue levels.

show abstract

mentioning

confidence: 99%

Technique to estimate human reaction time based on visual perception

Abbasi-Kesbi

Memarzadeh-Tehran

Deen

2017

Healthcare Technology Letters

View full text Add to dashboard Cite

show abstract

“…This application enables us to record egocentric video and sensor data simultaneously in a synchronized manner. The following types of sensor data are supported on the glass: accelerometer, gravity, gyroscope, linear acceleration, 1 Dataset:http://people.sutd.edu.sg/ 1000892/dataset arXiv:1601.06603v1 [cs.MM] 25 Jan 2016 magnetic field and rotation vector. These sensors are integrated in the device and they can help capturing accurate head motion of individuals when they are performing different activities.…”

Section: Multimodal Egocentric Activity Datasetmentioning

confidence: 99%

“…Research on automatic egocentric activity recognition has been focusing on using two broad categories of data: low-dimensional sensor data and high-dimensional visual data. Low-dimensional sensor data such as GPS, light, temperature, direction or accelerometer data has been found to be useful for activity recognition [1,2,3,4,5]. [2] proposes features for egocentric activity recognition computed from cell-phone accelerometer data.…”

Section: Introductionmentioning

confidence: 99%

Egocentric activity recognition with multimodal fisher vector

Song

Cheung

Chandrasekhar

et al. 2016

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

With the increasing availability of wearable devices, research on egocentric activity recognition has received much attention recently. In this paper, we build a Multimodal Egocentric Activity dataset which includes egocentric videos and sensor data of 20 fine-grained and diverse activity categories. We present a novel strategy to extract temporal trajectory-like features from sensor data. We propose to apply the Fisher Kernel framework to fuse video and temporal enhanced sensor features. Experiment results show that with careful design of feature extraction and fusion algorithm, sensor data can enhance information-rich video data. We make publicly available the Multimodal Egocentric Activity dataset to facilitate future research.

show abstract

“…Complementary to vision data, inertial sensor data (e.g., gyroscopes and accelerometers) provide position and direction information of the wearable device, which may facilitate human activity recognition for egocentric videos. Recently, with the advancement and application of wearable inertial sensors, multi-modal methods, i.e., combining vision data and sensor data to recognize human activities, are of widespread interest, which may promote vision-based methods [16], [17], [18]. Some pioneering work [17] uses LSTM to learn the feature from sensor data and CNNs to learn the feature from vision data, which are fused together to predict wearer's activity.…”

Section: Introductionmentioning

confidence: 99%

Towards Continual Egocentric Activity Recognition: A Multi-modal Egocentric Activity Dataset for Continual Learning

Xu¹,

Wu²,

Pan³

et al. 2023

Preprint

View full text Add to dashboard Cite

With the rapid development of wearable cameras, a massive collection of egocentric video for first-person visual perception becomes available. Using egocentric videos to predict first-person activity faces many challenges, including limited field of view (FoV), occlusions, and unstable motions. Observing that sensor data from wearable devices facilitates human activity recognition (HAR), activity recognition using multi-modal data is attracting increasing attention. However, the deficiency of related dataset hinders the development of multi-modal deep learning for egocentric activity recognition. Nowadays, deep learning in real world has led to a focus on continual learning that often suffers from catastrophic forgetting. But the catastrophic forgetting problem of continual learning for egocentric activity recognition, especially in the context of multiple modalities, remains unexplored due to unavailability of dataset. In order to assist this research, in this paper, we present a multi-modal egocentric activity dataset for continual learning named UESTC-MMEA-CL, which is collected by self-developed glasses integrating a firstperson camera and wearable sensors. It contains synchronized data of videos, accelerometers, and gyroscopes, for 32 types of daily activities, performed by 10 participants wearing the glasses. The collection device and process of our dataset are described. Its class types and scale are compared with other publicly available multi-modal datasets for egocentric activity recognition. The statistical analysis of the sensor data is given to show the auxiliary effects for different behaviors. And results of egocentric activity recognition are reported when using separately, and jointly, three modalities: RGB, acceleration, and gyroscope, on a base multi-modal network architecture. To explore the catastrophic forgetting in continual learning tasks on UESTC-MMEA-CL, four baseline methods are extensively evaluated with different multi-modal combinations. We hope the UESTC-MMEA-CL dataset can promote future studies on continual learning for firstperson activity recognition in wearable applications. Our dataset will be released soon.

show abstract

Implementation study of wearable sensors for activity recognition systems

Cited by 12 publications

References 20 publications

Technique to estimate human reaction time based on visual perception

Technique to estimate human reaction time based on visual perception

Egocentric activity recognition with multimodal fisher vector

Towards Continual Egocentric Activity Recognition: A Multi-modal Egocentric Activity Dataset for Continual Learning

Contact Info

Product

Resources

About