Filters








1,135 Hits in 7.7 sec

Detecting Transitions in Manual Tasks from Wearables: An Unsupervised Labeling Approach

Sebastian Böttcher, Philipp Scholl, Kristof Van Laerhoven
2018 Informatics  
Additionally, we tested the approach on a novel dataset in a manufacturing lab environment, describing an existing sequential manufacturing process.  ...  We investigate different clustering algorithms on wearable inertial sensor data recorded on par with video data, to automatically create transition marks between task steps.  ...  The CMU Kitchen dataset used for this research was obtained from http://kitchen.cs.cmu.edu/, and their data collection was funded in part by the National Science Foundation under Grant No.  ... 
doi:10.3390/informatics5020016 fatcat:x2sun6l5hfggbh5wlxhazf5lgu

Hidden Markov Model-Based Smart Annotation for Benchmark Cyclic Activity Recognition Database Using Wearables

Christine F. Martindale, Sebastijan Sprager, Bjoern M. Eskofier
2019 Sensors  
Using this pipeline, we have collected and labeled a dataset with over 150,000 labeled cycles, each with 2 phases, from 80 subjects, which we have made publicly available.  ...  The pipeline consists of three smart annotation approaches, namely edge detection of the pressure data, local cyclicity estimation, and iteratively trained hierarchical hidden Markov models.  ...  In future applications a semi-supervised or real time approach for task labeling could be implemented.  ... 
doi:10.3390/s19081820 fatcat:fgx4oszkpzhkhcfxbpbojqh6oi

Physical Human Activity Recognition Using Wearable Sensors

Ferhat Attal, Samer Mohammed, Mariam Dedabrishvili, Faicel Chamroukhi, Latifa Oukhellou, Yacine Amirat
2015 Sensors  
This comparison highlights which approach gives better performance in both supervised and unsupervised contexts.  ...  This paper presents a review of different classification techniques used to recognize human activities from wearable inertial sensor data.  ...  The acquired data were manually labeled by an independent operator.  ... 
doi:10.3390/s151229858 pmid:26690450 pmcid:PMC4721778 fatcat:qx33pz7hsvau5pvfsbou7vvc3m

Ajalon: Simplifying the authoring of wearable cognitive assistants

Truong An Pham, Junjue Wang, Roger Iyengar, Yu Xiao, Padmanabhan Pillai, Roberta Klatzky, Mahadev Satyanarayanan
2021 Software, Practice & Experience  
Wearable Cognitive Assistance (WCA) amplifies human cognition in real time through a wearable device and low-latency wireless access to edge computing infrastructure.  ...  It is inspired by, and broadens, the metaphor of GPS navigation tools that provide real-time step-by-step guidance, with prompt error detection and correction.  ...  Data labeling from OpenTPOD: Essentially, OpenTPOD halved the time for the data-labeling step in building the expected-object detectors.  ... 
doi:10.1002/spe.2987 fatcat:qrn2icargjakfasl5b5duk5dpy

Transitional Activity Recognition with Manifold Embedding

Raza Ali, Louis Atallah, Benny Lo, Guang-Zhong Yang
2009 2009 Sixth International Workshop on Wearable and Implantable Body Sensor Networks  
To observe, as well as quantify, transitional activities, a manifold embedding approach is proposed in this paper.  ...  The method uses a spectral graph partitioning and transition labelling approach for identifying principal and transitional activity patterns.  ...  vertices T Figure 3 . 3 An example of marker inconsistency during manual annotation in situ, showing the accuracy of the proposed method in detecting activity transitions.  ... 
doi:10.1109/bsn.2009.42 dblp:conf/bsn/AliALY09 fatcat:5ay4xucmbvbcllzzrc6xw54goe

In the sight of my wearable camera: Classifying my visual experience [article]

Alessandro Perina, Nebojsa Jojic
2013 arXiv   pre-print
early detection of dementia to everyday use of wearable camera streams for automatic reminders, and visual stream exchange.  ...  over dozens of visual scenes (locations) encountered over the course of several weeks of a human life with accuracy of over 80%, and this opens up possibility for numerous novel vision applications, from  ...  early detection of dementia to everyday use of wearable camera streams for automatic reminders, and visual stream exchange.  ... 
arXiv:1304.7236v1 fatcat:ol2bnkbktffplmy2cd2do4gm4q

Deep Auto-Set: A Deep Auto-Encoder-Set Network for Activity Recognition Using Wearables [article]

Alireza Abedin Varamin, Ehsan Abbasnejad, Qinfeng Shi, Damith Ranasinghe, Hamid Rezatofighi
2018 arXiv   pre-print
Automatic recognition of human activities from time-series sensor data (referred to as HAR) is a growing area of research in ubiquitous computing.  ...  Most recent research in the field adopts supervised deep learning paradigms to automate extraction of intrinsic features from raw signal inputs and addresses HAR as a multi-class classification problem  ...  As opposed to other domains (e.g. image recognition) where human visualization of raw data alleviates the labeling process, manual annotation of sensor signals is a tedious task.  ... 
arXiv:1811.08127v1 fatcat:ryuljwltizglbkx7pxjxy4kwyu

Easing Power Consumption of Wearable Activity Monitoring with Change Point Detection

Culman, Aminikhanghahi, J. Cook
2020 Sensors  
CPAM employs unsupervised change point detection to detect likely activity transition times.  ...  We validate our approach using smartwatch data collected and labeled by 66 subjects.  ...  However, their work relied on manual identification of activity transitions. We replace this step with automatic transition labeling via change point detection.  ... 
doi:10.3390/s20010310 pmid:31935907 pmcid:PMC6982794 fatcat:6swa7oz76be6pf3aeiupe72whq

Recognizing Activities and Spatial Context Using Wearable Sensors [article]

Amarnag Subramanya, Alvin Raj, Jeff A. Bilmes, Dieter Fox
2012 arXiv   pre-print
A key goal, however, in designing our overall system is to be able to perform accurate inference decisions while minimizing the amount of hardware an individual must wear.  ...  GPS measurements, and measurements from a small mountable sensor board.  ...  Additional thanks go to the Intel Research Lab in Seattle for providing the sensor boards used in this research.  ... 
arXiv:1206.6869v1 fatcat:xty2yyvionakzjfimoet5pdvpe

Wearable-Sensor-based Detection and Prediction of Freezing of Gait in Parkinson's Disease: A Review

Pardoel, Kofman, Nantel, Lemaire
2019 Sensors  
Details were extracted regarding participants, walking task, sensor type and body location, detection or prediction approach, feature extraction and selection, classification method, and detection and  ...  This review searched the Scopus, PubMed, and Web of Science databases to identify studies that used wearable sensors to detect or predict FOG episodes in Parkinson's disease.  ...  Unsupervised FOG detection approaches are appealing since they do not require data labelling; however, few studies have used unsupervised FOG detection, and unsupervised models performance has been worse  ... 
doi:10.3390/s19235141 pmid:31771246 pmcid:PMC6928783 fatcat:q2m6pxtvpjb6zpthyo4xkjsg4y

Estimating Attention in Exhibitions Using Wearable Cameras

Ali S. Razavian, Omid Aghazadeh, Josephine Sullivan, Stefan Carlsson
2014 2014 22nd International Conference on Pattern Recognition  
Our method is a combination of multiple state of the art techniques from different vision tasks such as tracking, image matching and retrieval.  ...  This paper demonstrates a system for automatic detection of visual attention and identification of salient items at exhibitions (e.g. museum or an auction).  ...  METHOD Our work aims to identify interesting items in an exhibition from several videos taken by a head mounted wearable camera.  ... 
doi:10.1109/icpr.2014.465 dblp:conf/icpr/RazavianASC14 fatcat:6d3dascszva6znn2vsi2akvh7e

Discovering Characteristic Actions from On-Body Sensor Data

David Minnen, Thad Starner, Irfan Essa, Charles Isbell
2006 Wearable Computers (ISWC), Proceedings of the IEEE International Symposium on  
We present an approach to activity discovery, the unsupervised identification and modeling of human actions embedded in a larger sensor stream.  ...  Rather than learn models from hand-labeled sequences, we attempt to discover motifs, sets of similar subsequences within the raw sensor stream, without the benefit of labels or manual segmentation.  ...  Traditional arguments for unsupervised learning, such as reducing cost by precluding the need to manually label data, aiding adaptation to non-stationary patterns, and providing early exploratory tools  ... 
doi:10.1109/iswc.2006.286337 dblp:conf/iswc/MinnenSEI06 fatcat:iispfr6wqfbyrjll66uebwgl5i

Deep Physiological Arousal Detection in a Driving Simulator Using Wearable Sensors

Aaqib Saeed, Stojan Trajanovski, Maurice van Keulen, Jan van Erp
2017 2017 IEEE International Conference on Data Mining Workshops (ICDMW)  
The physiological signals are collected from 11 participants by wrist wearable devices.  ...  For this purpose, we used deep learning algorithms to detect arousal level, namely, under-aroused, normal and overaroused for professional truck drivers in a simulated environment.  ...  Problem Definition We considered the problem of arousal detection as a supervised sequence classification. In this task, the objective is to assign a single label to an input sequence.  ... 
doi:10.1109/icdmw.2017.69 dblp:conf/icdm/SaeedTKE17 fatcat:vvsv22kd5vhzpgdy6xu776lfsm

Deep Auto-Set

Alireza Abedin Varamin, Ehsan Abbasnejad, Qinfeng Shi, Damith C. Ranasinghe, Hamid Rezatofighi
2018 Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services - MobiQuitous '18  
Automatic recognition of human activities from time-series sensor data (referred to as HAR) is a growing area of research in ubiquitous computing.  ...  Most recent research in the field adopts supervised deep learning paradigms to automate extraction of intrinsic features from raw signal inputs and addresses HAR as a multi-class classification problem  ...  As opposed to other domains (e.g. image recognition) where human visualization of raw data alleviates the labeling process, manual annotation of sensor signals is a tedious task.  ... 
doi:10.1145/3286978.3287024 dblp:conf/mobiquitous/VaraminASRR18 fatcat:zjtalwqusbbpvmm47frgqucfnu

Wearable-Based Affect Recognition—A Review

Philip Schmidt, Attila Reiss, Robert Dürichen, Kristof Van Laerhoven
2019 Sensors  
Affect recognition is an interdisciplinary research field bringing together researchers from natural and social sciences.  ...  However, existing literature surveys lack a comprehensive overview of state-of-the-art research in wearable-based affect recognition.  ...  Customisation could happen, for instance, via an active labelling approach, where the user is occasionally asked to provide labels.  ... 
doi:10.3390/s19194079 pmid:31547220 pmcid:PMC6806301 fatcat:axkkxpscangaxf3dbvxighvfme
« Previous Showing results 1 — 15 out of 1,135 results