On-body localization of wearable devices: An investigation of position-aware activity recognition
The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Abstract
Human activity recognition using mobile device sensors is an active area of research in pervasive computing. In our work, we aim at implementing activity recognition approaches that are suitable for real life situations. This paper focuses on the problem of recognizing the on-body position of the mobile device which in a real world setting is not known a priori. We present a new real world data set that has been collected from 15 participants for 8 common activities were they carried 7 wearable devices in different positions. Further, we introduce a device localization method that uses random forest classifiers to predict the device position based on acceleration data. We perform the most complete experiment in on-body device location that includes all relevant device positions for the recognition of a variety of different activities. We show that the method outperforms other approaches achieving an F-Measure of 89% across different positions. We also show that the detection of the device position consistently improves the result of activity recognition for common activities.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The performance of supervised learning methods is particularly dependent on large-scale labeled datasets, which require significant human efforts and financial costs [62].”
From Spatial-Temporal Masked Autoencoder for Multi-Device Wearable Human Activity Recognition (2023) · cites this paper as [62]
Citations
Cites
Breiman, L. (2001) · Machine Learning · doi:10.1023/a:1010933404324 · Cited by 133,538 · cited 1×
“Random forest classifiers are one of these ensemble methods that have been proposed by Breimann [17].”
MethodThe random subspace method for constructing decision forests
Ho, T. K. (1998) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.709601 · Cited by 6,842 · cited 1×
“This is necessary to ensure that the different generated decision trees are uncorrelated [18].”
BackgroundActivity Recognition from User-Annotated Acceleration Data
Bao, L., & Intille, S. S. (2004) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-24646-6_1 · Cited by 3,147 · cited 2×
“The development of wearable devices such as smart-phones, smart-watches and fitness wristbands feature a variety of sensors that are carried all day long by many people (compare [3]) provide new opportunities for continuous monitoring of human activities such as running, sitting, standing, or walking [5].”
BackgroundActivity recognition using cell phone accelerometers
Kwapisz, J. R., Weiss, G. M., & Moore, S. A. (2011) · ACM SIGKDD Explorations Newsletter · doi:10.1145/1964897.1964918 · Cited by 2,912 · cited 2×
“The sampling rate was chosen with consideration of battery life as well as with reference to previous studies [7], [14].”
BackgroundMethodA Survey on Human Activity Recognition using Wearable Sensors
Lara, O. D., & Labrador, M. A. (2013) · IEEE Communications Surveys & Tutorials · doi:10.1109/surv.2012.110112.00192 · Cited by 2,662 · cited 1×
“Preceding studies in the context of activity recognition already examined different settings regarding the window size and meaningful features [15].”
BackgroundActivity recognition from accelerometer data
Ravi, N., Dandekar, N., Mysore, P., & Littman, M. L. (2005) · Innovative Applications of Artificial Intelligence · Cited by 1,775
Preece, S. J., Goulermas, J. Y., Kenney, L. P. J., & Howard, D. (2009) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2008.2006190 · Cited by 551 · cited 2×
“Nevertheless, some researchers have compared different groups of features and also stated that frequency-based features improve the accuracy of the recognition [16].”
BackgroundSensor Positioning for Activity Recognition Using Wearable Accelerometers
Atallah, L., Lo, B., King, R., & Yang, G.-Z. (2011) · IEEE Transactions on Biomedical Circuits and Systems · doi:10.1109/tbcas.2011.2160540 · Cited by 346 · cited 5×
“Further studies have shown that the optimal sensor placement depends on the activity to be recognized [6].”
BackgroundResultInfluentialYang, J.-Y., Wang, J.-S., & Chen, Y.-P. (2008) · Pattern Recognition Letters · doi:10.1016/j.patrec.2008.08.002 · Cited by 335 · cited 1×
“A similar distinction has been made in [19] to improve the accuracy of activity recognition.”
BackgroundSun, L., Zhang, D., Li, B., Guo, B., & Li, S. (2010) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-16355-5_42 · Cited by 251 · cited 2×
“Activity Recognition is an active field of research in pervasive computing [1]–[4].”
Background
Cited by
Items 201–300 of 509
FedIERF: Federated Incremental Extremely Random Forest for Wearable Health Monitoring
Hu, C., Hu, L., Yuan, L., Lu, D., Lyu, L., & Chen, Y. (2023) · Journal of Computer Science and Technology · doi:10.1007/s11390-023-3009-0 · Cited by 11
Wang, T., Chen, Y., Yang, Q., Sun, D., Wang, R., Li, J., Kimura, T., & Abdelzaher, T. (2024) · International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn61486.2024.10637566 · Cited by 11
“RealWorld HAR [27] The RealWorld HAR dataset includes data collected from 15 participants engaged in five dynamic activities, including walking, running, ascending stairs, descending stairs, and jumping.”
MethodWen, X., Wang, Y., Yuan, M., Geng, Y., Yu, H., & Ge, Z. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3465593 · Cited by 11
“The RealWorld dataset comprises 18 h of data recorded from 15 participants using Samsung Galaxy S4 smartphones and LG G Watch R smartwatches, each set to a sampling rate of 50 Hz [36].”
Abnormality Detection Approach using Deep Learning Models in Smart Home Environments
Sukor, A. S. A., Zakaria, A., Rahim, N. A., Kamarudin, L. M., & Nishizaki, H. (2019) · International Conference on Communications and Broadband Networking (ICCBN) · doi:10.1145/3330180.3330185 · Cited by 11
Hand with Sensing Sphere: Body-Centered Spatial Interactions with a Hand-Worn Spherical Camera
Arakawa, R., Maekawa, A., Kashino, Z., & İnami, M. (2020) · Symposium on Spatial User Interaction · doi:10.1145/3385959.3418450 · Cited by 11
“For example, such information could help to develop medical monitoring systems [55], improving the quality of activity recognition [53], and automate calibration processes [43].”
BackgroundHuman Activity Recognition by Deep Convolution Neural Networks and Principal Component Analysis
Aljarrah, A. A., & Ali, A. H. (2021) · Intelligent systems reference library · doi:10.1007/978-3-030-57835-0_10 · Cited by 11
Exploring a multi-sensor picking process in the future warehouse
Diete, A., Weiland, L. R., Sztyler, T., & Stuckenschmidt, H. (2016) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/2968219.2968270 · Cited by 11
“In this context, recently, researchers investigated the recognition of common postures such as standing and walking in a real-world scenario relying on an acceleration sensor and considering several on-body position of the wearable device [8].”
BackgroundMethodSelf-citationCo-MEAL: Cost-Optimal Multi-Expert Active Learning Architecture for Mobile Health Monitoring
Saeedi, R., Sasani, K., & Gebremedhin, A. H. (2017) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3107411.3107430 · Cited by 11
“The dataset we used is publicly available at Data and Web Science Research Group, University of Mannheim [24].”
MethodMoves like Jagger: Exploiting variations in instantaneous gait for spontaneous device pairing
Schürmann, D., Brüsch, A., Nguyen, L. N., Sigg, S., & Wolf, L. C. (2018) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2018.03.006 · Cited by 11
“In particular, we utilize the Mannheim dataset presented in detail in [32] for the use in position aware activity recognition.”
MethodSlopewise Aggregate Approximation SAX: keeping the trend of a time series
Pappa, L., Karvelis, P. S., Georgoulas, G., & Stylios, C. D. (2021) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci50451.2021.9660130 · Cited by 11
Federated Learning for Time-Series Healthcare Sensing with Incomplete ModalitiesPreprint
Orzikulova, A., Kwak, J., Shin, J., & Lee, S.-J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2405.11828 · Cited by 11
“RealWorld HAR [48] is a multi-device multimodal dataset for human activity recognition (HAR), featuring data from fifteen participants engaged in eight activities.”
Domain Adaptation for Inertial Measurement Unit-based Human Activity Recognition: A SurveyPreprint
Chakma, A., Faridee, A. Z. M., Ghosh, I., & Roy, N. (2023) · arXiv (Cornell University) · doi:10.13016/m2pk2o-njvc · Cited by 11
Investigating (re)current state-of-the-art in human activity recognition datasets
Bock, M., Hoelzemann, A., Moeller, M., & Laerhoven, K. V. (2022) · Frontiers in Computer Science · doi:10.3389/fcomp.2022.924954 · Cited by 10
Yussif, S. B., Xie, N., Yang, Y., Huang, Y., Wang, G., & Du, Z. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3446290 · Cited by 10
“The sections detail the WISDM [44], UCI-HAR [45], PAMAP2 [46], RWHAR [47], and USCHAD [48] datasets used to evaluate the effectiveness of our proposed method.”
MethodExploring Semi-Supervised Methods for Labeling Support in Multimodal Datasets
Diete, A., Sztyler, T., & Stuckenschmidt, H. (2018) · Sensors · doi:10.3390/s18082639 · Cited by 10
“Thus, we enhanced an Android application of a previous work [24] where especially the support of smartglasses was added.”
MethodSelf-citationSaeedi, R., Sasani, K., & Gebremedhin, A. H. (2020) · Sensors · doi:10.3390/s20071932 · Cited by 10
“This dataset concerns daily activities and is publicly available at the Web Research Group at University of Mannheim [50].”
MethodDeep Architectures for Human Activity Recognition using Sensors
Baloch, Z., Shaikh, F. K., & Unar, M. A. (2019) · 3C Tecnología_Glosas de innovación aplicadas a la pyme · doi:10.17993/3ctecno.2019.specialissue2.14-35 · Cited by 10
“…(Goldberger, et al., 2000) EEG, EOG, EMG Sleep stages i.e. Awake, Stage–1, Stage 2, Stage 3, Stage 4, REM 100 –– 20 Sleep Analysis RealWorld HAR (Sztyler & Stuckenschmidt, 2016) A, G, GPS, L, MF, S Climbing Downstairs, ClimbingUpstairs, Jumping, Lying, Standing, Sitting,…”
INIM: Inertial Images Construction with Applications to Activity Recognition
Daniel, N., & Klein, I. (2021) · Sensors · doi:10.3390/s21144787 · Cited by 10
“SLR [36] HAR1 [33] HAR2 [34] HAR3 [35] HAR4 [35]”
InfluentialSensiX: A System for Best-effort Inference of Machine Learning Models in Multi-device Environments
Min, C., Mathur, A., Montanari, A., & Kawsar, F. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3173914 · Cited by 10
“For the additional details, refer to [11], [38] for the architecture of sensing models and [29], [37] for the datasets.”
BackgroundMethodTransformer Networks for Data Augmentation of Human Physical Activity RecognitionPreprint
Ramachandra, S., Hoelzemann, A., & Laerhoven, K. V. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.01081 · Cited by 10
“We validated our network with leave-one-subject-out (LOSO) cross-Validation and on two publicly available datasets, PAMAP2 [10] and Real World Human Activity Recognition (RWHAR) [13].”
MethodInfluentialHow Much Unlabeled Data is Really Needed for Effective Self-Supervised Human Activity Recognition?
Dhekane, S. G., Haresamudram, H., Thukral, M., & Plötz, T. (2023) · Proceedings of the 2023 International Symposium on Wearable Computers · doi:10.1145/3594738.3611366 · Cited by 9
“For HHAR, PAMAP2, and RealWorld, 296, 354, and 1,204 minutes of data represent the maximum amount of data that can be used to perform both pre-training and fine-tuning.”
InfluentialFederated Quantum Kernel-Based Long Short-Term Memory for Human Activity Recognition
Hsu, Y.-C., Jiang, J.-C., Lin, C.-H., Chen, W.-T., Peng, K. C., Tiwari, P., Chen, S. Y.-C., & Kuo, E.-J. (2025) · IEEE International Conference on Quantum Computing and Engineering (QCE) · doi:10.1109/qce65121.2025.10293 · Cited by 9
“Formal mathematical formulation of QK-LSTM Cell is given by: where: quantum kernel functions tailored to the respective gates.”
Potter, V., Tran, H., Mobley, D. R., Bertisch, S. M., John, D., & Intille, S. S. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3770639 · Cited by 9
Oishi, N., Birch, P. M., Roggen, D., & Lago, P. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3661047 · Cited by 9
“REALWORLD [36] includes IMU and video recordings from 15 participants performing eight activities: walking, running, sitting, standing, lying, climbing stairs up, climbing stairs down, and jumping, without a null class.”
Wonderwall: A Virtual-to-Real Foundation Model for IMU-based HAR
Miao, S., & Chen, L. (2026) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3789688 · Cited by 9
“…2 Pretraining Datasets: We selected eight datasets based on comprehensive coverage of diverse body-mounted sensor positions and representation of fundamental physical activities: USCHAD [68], Harth [31], RealWorld [49], Opportunity [45], HAR70+ [51], MotionSense [32], TNDA-HAR [59], and UCIHAR [3].”
Chen, Y., Hu, C., Kimura, T., Li, Q., Liu, S., Wu, F., & Chen, G. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3699779 · Cited by 9
“As a reference, the included time-domain augmentations are scaling, permutation, negation, time warp, magnitude warp, horizontal flip, jitter, channel shuffle, and time masking.”
Towards real world activity recognition from wearable devices
Sztyler, T. (2017) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomw.2017.7917535 · Cited by 9
“Therefore, we examined [1] its real world feasibility by considering all relevant on-body positions and common physical activities.”
BackgroundSelf-citationWireless non-invasive motion tracking of functional behavior
Zhang, H., Smeddinck, J. D., Malaka, R., Shu, Y., Chen, C., He, B., Fu, Z., & Lawo, M. (2019) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2019.01.006 · Cited by 9
“Sztyler et al. [28] present a new real-world data set that has been collected from 15 participants for eight common activities where they carried seven wearable devices in different positions.”
BackgroundPappa, L., Karvelis, P. S., Georgoulas, G., & Stylios, C. D. (2020) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci47803.2020.9308497 · Cited by 9
“The proposed method is tested with signals from a publicly available web database [16, 17] that contains triaxial linear acceleration and angular velocity signals recorded from the accelerometer and the gyroscope of the device at a sampling rate of 50 Hz.”
BackgroundMethodLiang, K.-Y., Srinivasan, A., & Andresen, J. C. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9892377 · Cited by 9
“1https://www.cs.toronto.edu/ kriz/cifar.html 2https://cs.stanford.edu/ acoates/stl10/ b) Human activity classification: HAPT3 [15], WISDM4 [16], and RealWorld HAR5 [17] (RWHAR) are three different human activity recognition datasets, in which they have 5 labels in common.”
BackgroundUniMTS: Unified Pre-training for Motion Time Series
Zhang, X., Teng, D., Chowdhury, R., Li, S., Hong, D., Gupta, R., & Shang, J. (2024) · Advances in Neural Information Processing Systems 37 · doi:10.52202/079017-3414 · Cited by 8
Haresamudram, H., Beedu, A., Rabbi, M., Saha, S., Essa, I., & Ploetz, T. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i1.32004 · Cited by 8
Bukit, T. A., Pillado, E. P. B., Yahya, B. N., & Lee, S. (2025) · Applied Soft Computing · doi:10.1016/j.asoc.2025.113793 · Cited by 8
No One Size (PPM) Fits All: Towards Privacy in Stream Processing Systems
Fomichev, M., Luthra, M., Benndorf, M., & Agnihotri, P. (2023) · ACM International Conference on Distributed and Event-Based Systems (DEBS) · doi:10.1145/3583678.3596889 · Cited by 8
“evaluate ObscureNet on the RealWorld (HAR) dataset that contains IMU data from five devices located as such: head, upper arm, waist, thigh, and shin [26]; originally, ObscureNet is validated on the IMU data coming from a single thigh-worn device [11].”
BackgroundWang, H., Wang, X., Lu, C., Yuan, M., Wang, Y., Yu, H., & Li, H. (2024) · IEEE Access · doi:10.1109/access.2024.3422256 · Cited by 8
“Sensor displacement occurs when sensors are not in their ideal or predefined locations, which can involve rotation, translation, or both [4], [5], [6].”
BackgroundCHAR: Composite Head-Body Activities Recognition With a Single Earable Device
Zhu, P., Zhu, Y., Li, W., He, Y., Zou, Y., Wu, K., & Leung, V. C. M. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3548647 · Cited by 8
“In addition to the self-collected dataset, we also used a public dataset RealWorld [42].”
FedOpenHAR: Federated Multitask Transfer Learning for Sensor-Based Human Activity Recognition
İşgüder, E., & İncel, Ö. D. (2025) · Journal of Computational Biology · doi:10.1089/cmb.2024.0631 · Cited by 8
“They also perform experiments and report their results using the CIFAR-10 dataset.”
Kraft, D., Srinivasan, K., & Bieber, G. (2020) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3389189.3397983 · Cited by 8
“To enrich the Notch, Smartwatch and SmartFall datasets with additional activities of daily living, we added 500 random 10 second segments from the RealWorld Human Activity Recognition [43] dataset to Smartfall, Smartwatch and Notch dataset.”
Method“Chatty Devices” and edge-based activity classification
Lakoju, M., Javed, A., RANA, O. F., Burnap, P., Atiba, S., & Cherkaoui, S. (2021) · Discover Internet of Things · doi:10.1007/s43926-021-00004-9 · Cited by 8
“Human Activity Recognition Dataset The publicly available Human Activity Recognition (HAR) data set [46] covers acceleration, GPS, gyroscope, light, magnetic field, and sound level data of the activities.”
BackgroundConstruction Tasks Electronic Process Monitoring: Laboratory Circuit-Based Simulation Deployment
Calvetti, D., Sanhudo, L., Mêda, P., Martins, J. P., Gonçalves, M. C., & Sousa, H. J. C. D. (2022) · Buildings · doi:10.3390/buildings12081174 · Cited by 8
Incremental Data Stream Classification with Adaptive Multi-Task Multi-View Learning
Wang, J., Shi, M., Zhang, X., Li, Y., Yuan, Y., Yang, C., & Yu, D. (2023) · Big Data Mining and Analytics · doi:10.26599/bdma.2023.9020006 · Cited by 8
Lightweight Modeling of User Context Combining Physical and Virtual Sensor Data
Campana, M. G., Chatzopoulos, D., Delmastro, F., & Hui, P. (2018) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3267305.3274178 · Cited by 8
“However, most of the publicly available datasets (e.g., UMA Fall [35], UniMiB SHAR [15], and RealWorld (HAR) [36]) take into account only a limited number of sensors since their main purpose is to assist the creation of models for specific activity recognition (e.g., user’s gait and fall detection).”
BackgroundRokni, S. A., & Ghasemzadeh, H. G. (2019) · ACM Transactions on Design Automation of Electronic Systems · doi:10.1145/3318044 · Cited by 8
“Although prior studies combine sensor localization or context detection within activity classification [24, 31], they require a set of labeled data to train the localization/context-learning model.”
BackgroundActivity Recognition in the City using Embedded Systems and Anonymous Sensors
Baghezza, R., Bouchard, K., Bouzouane, A., & Gouin-Vallerand, C. (2020) · Procedia Computer Science · doi:10.1016/j.procs.2020.03.140 · Cited by 8
“The position of the device itself can affect the activity recognition results as well[15].”
BackgroundA study on smartphone sensor-based Human Activity Recognition using deep learning approaches
Mondal, R., Mukhopadhyay, D., Barua, S., Singh, P., Sarkar, R., & Bhattacharjee, D. (2021) · Elsevier eBooks · doi:10.1016/b978-0-12-822260-7.00006-6 · Cited by 8
Pervasive Pose Estimation for Fall Detection
Luo, J., Bai, R., He, S., & Shin, K. G. (2022) · ACM Transactions on Computing for Healthcare · doi:10.1145/3478027 · Cited by 8
“The body part where the device is to be attached/placed [11, 27] has been identified and the effect of sensor placement on detection [1, 2] has been investigated further.”
BackgroundImproving Motion-based Activity Recognition with Ego-centric Vision
Diete, A., Sztyler, T., Weiland, L. R., & Stuckenschmidt, H. (2018) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomw.2018.8480334 · Cited by 8
“In contrast to our previous work on activity recognition (compare [2]), we do not only use features from motion sensors, but add features extracted from the ego-”
BackgroundSelf-citationLiu, M., Xiang, D., Cheng, X., Liu, X., Zhang, D., Chen, S., & Jensen, C. S. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2402.09434 · Cited by 8
“• RealWorld: Described in [36], this dataset collects activities from 15 users.”
BackgroundFederated learning for next generation intelligent applications
Khan, A. R. (2025) · Enlighten: Theses (The University of Glasgow) · doi:10.5525/gla.thesis.85028 · Cited by 8
WHAR Datasets: An Open Source Library for Wearable Human Activity Recognition
Burzer, M., King, T., Riedel, T., Beigl, M., & Röddiger, T. (2025) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3714394.3756254 · Cited by 7
Gait Device Pairing in Unconstrained Environments With Signal Morphing
Boshoff, D., Nkrow, R. E., Silva, B., & Hancke, G. P. (2024) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2024.3405976 · Cited by 7
“Popular datasets utilized by gait key-sharing papers [25], [34], [36] include RealWorld (HAR) [39] as well as the dataset used by Xu et al. [21], [22].”
Eff-WHAR: A Lightweight Design for Efficient Wearable Sensor-Based Human Activity Recognition
Chen, Z., Zou, H., Wang, L., Wang, B., Zhang, F., Ma, S., Pan, Y., & Li, J. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3509961 · Cited by 8
Sarker, P., Tiang, J. J., & Nahid, A.-A. (2025) · Sensors · doi:10.3390/s25175303 · Cited by 7
“…smartphones have made them ideal devices for data acquisition using their accelerometer and gyroscope sensors, as seen in the UCI HAR dataset [4], WISDM HAR [5], and RealWorld HAR [6], among others, as well as “KU-HAR: An Open Dataset for Human Activity Recognition”, a similar type of dataset [7].”
Exploring Activity Recognition in Multi-device Environments using Hierarchical Federated Learning
Sandhu, M. M., Silvera‐Tawil, D., Lü, W., Borges, P., & Kusý, B. (2024) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops59983.2024.10503023 · Cited by 7
“To evaluate the proposed mechanism, we employed Real-World [12], a publicly available activity recognition dataset.”
MethodShuffleFL: Addressing Heterogeneity in Multi-Device Federated Learning
Zhu, R., Yang, M., & Wang, Q. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3659621 · Cited by 7
“To assess the effectiveness of ShuffleFL in handling data inconsistency, specifically non-IID input spaces introduced by different devices, we employ a real-world multi-device dataset RealWorld [50] for human activity recognition.”
MethodA multi-sensor algorithm for activity and workflow recognition in an industrial setting
Thomay, C., Gollan, B., Haslgrübler, M., Ferscha, A., & Heftberger, J. (2019) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3316782.3321523 · Cited by 7
“When considering the employed sensors, a major focus has been put on wearable, (i) explicitly body-worn sensor setups [29], with special focus on IMU sensor types [5], as well as (ii) smartphones [6], [18],[11], or (iii) smart watches [27] for activity assessment.”
BackgroundEk, S., Portet, F., Lalanda, P., & Vega, G. (2021) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops51409.2021.9431080 · Cited by 7
Deep Learning-Based Human Posture Recognition
Ayre-Storie, A., & Zhang, L. (2021) · International Conference on Machine Learning and Cybernetics · doi:10.1109/icmlc54886.2021.9737241 · Cited by 7
“The datasets used for testing include the RealWorld (HAR) by the University of Mannheim [16]; the Multiple Pose Human Body Database [17]; the Standford 40 Action Dataset [18] and the Northwestern-UCLA Multiview Action 3D Dataset [19].”
MethodPosition Invariance for Wearables: Interchangeability and Single-Unit Usage via Machine Learning
Yurtman, A., Barshan, B., & Redif, S. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3044754 · Cited by 7
Conformance Constraint Discovery: Measuring Trust in Data-Driven SystemsPreprint
Fariha, A., Tiwari, A., Radhakrishna, A., Gulwani, S., & Meliou, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.01289 · Cited by 7
Tutorial on Deep Learning for Human Activity RecognitionPreprint
Bock, M., Hoelzemann, A., Moeller, M., & Laerhoven, K. V. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.06663 · Cited by 7
ColloSSL: Collaborative Self-Supervised Learning for Human Activity RecognitionPreprint
Jain, Y., Tang, C. I., Min, C., Kawsar, F., & Mathur, A. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2202.00758 · Cited by 7
“In the remaining three cases, ColloSSL shows comparable 𝐹1 score with 25% of labeled data when evaluated at the thigh-worn device in RealWorld (ColloSSL: 0.690, Supervised-single: 0.701), a higher 𝐹1 score with 100% of labeled data when evaluated at the ankle-worn device in PAMAP2 (ColloSSL: 0.784, Supervised-single:…”
MethodInfluentialEk, S., Wang, K., Portet, F., Lalanda, P., & Cao, J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2411.10595 · Cited by 7
“• RealWorld dataset [52] shows a data heterogeneous learning environment caused by differences in the location of the on-body wearable sensor.”
Cheng, C.-F., Lin, C. J., & Hu, Q.-X. (2025) · Artificial intelligence for engineering design analysis and manufacturing · doi:10.1017/s0890060424000313 · Cited by 6
“Moreover, most existing public databases are based on ADL (Anguita et al., 2013; Banos et al., 2014; Chavarriaga et al., 2013; Micucci et al., 2017; Reiss and Stricker, 2012; Stisen et al., 2015; Sztyler and Stuckenschmidt, 2016).”
OpenMAE: Efficient Masked Autoencoder for Vibration Sensing with Open-domain Data Enrichment
Hu, C., Chen, Y., Kara, D., Liu, S., Abdelzaher, T., Wu, F., & Chen, G. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3729485 · Cited by 6
MODA: Motion-Drift Augmentation for Inertial Human Motion Analysis
Wu, Y., Guo, S., & Qin, Y. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.02586 · Cited by 6
Hi-OSCAR: Hierarchical Open-set Classifier for Human Activity Recognition
McCarthy, C., Quirijnen, L., van Zandwijk, J. P., Geradts, Z., & Worring, M. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3770681 · Cited by 6
“HAC takes a bottom up approach, each class starts in its own cluster and pairs of clusters are merged until all classes have been combined into the Root.”
A Federated Learning Approach to Minimize Communication Rounds Using Noise Rectification
Mishra, R., & Gupta, H. P. (2024) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc57260.2024.10570893 · Cited by 7
“Finally, we implement the approach and perform experiments to evaluate its performance on existing datasets [9]–[11] and compare with state-of-the-art techniques [5], [12]–[14].”
MethodHuang, Y., Zhao, H., Zhou, Y., Riedel, T., & Beigl, M. (2024) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-031-63992-0_2 · Cited by 6
CARL: a running recognition algorithm for free-living accelerometer data
Davis, J. J., Strączkiewicz, M., Harezlak, J., & Gruber, A. H. (2021) · Physiological Measurement · doi:10.1088/1361-6579/ac41b8 · Cited by 6
“…data from eight previously-published activity recognition datasets which included running alongside other activities ( Reiss and Stricker 2012, Leutheuser et al 2013, Banos et al 2014, Sztyler and Stuckenschmidt 2016, Mannini et al 2017, Sucerquia et al 2017, John et al 2019, Weiss et al 2019 ) .”
CROMOSim: A Deep Learning-based Cross-modality Inertial Measurement Simulator
Hao, Y., Wang, B., & Zheng, R. (2022) · 2022 2nd International Workshop on Cyber-Physical-Human System Design and Implementation (CPHS) · doi:10.1109/cphs56133.2022.9804560 · Cited by 6
Open-Source Data Collection for Activity Studies at Scale
Hoelzemann, A., Pithan, J. S., & Laerhoven, K. V. (2022) · Smart innovation, systems and technologies · doi:10.1007/978-981-19-0361-8_2 · Cited by 6
Human Activity Recognition: Trends and Challenges
Thakur, D., & Pal, A. (2024) · Internet of things · doi:10.1007/978-3-031-60027-2_8 · Cited by 6
Pattern Recognition of Human Postures Using the Data Density Functional Method
Huang, S.-J., Wu, C.-J., & Chen, C.-C. (2018) · Applied Sciences · doi:10.3390/app8091615 · Cited by 6
“Among these applications, the hardware schemes of data acquirement commonly utilized wireless sensors [14–17], personal portable devices [9,16,18], robotic assistance [19–22], or other commercial equipment [1,23,24].”
MethodAdvancing Location-Invariant and Device-Agnostic Motion Activity Recognition on Wearable DevicesPreprint
Adaimi, R., Bedri, A., Gong, J., Kang, R., Arreaza-Taylor, J., Pascual, G.-M., Ralph, M., & Laput, G. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2402.03714 · Cited by 6
“The most closely related dataset is the RealWorld dataset [37], introduced for studying on-body localization of wearable devices.”
BackgroundLoRA-Edge: Tensor-Train–Assisted LoRA for Practical CNN Fine-Tuning on Edge Devices
Kwak, H., Lee, K., Lee, J., & Lee, W. (2026) · Design, Automation, and Test in Europe · doi:10.23919/date69613.2026.11539617 · Cited by 5
Experience Paper: Adopting Activity Recognition in On-demand Food Delivery Business
Xu, H., Zhang, Y., Gao, W., Shen, G., & Li, M. (2025) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/3680207.3765261 · Cited by 5
“While numerous public IMU datasets exist [28, 31, 35, 40, 41, 43] for activity recognition, none, to the best of our knowledge, address such a level of diversity required for large-scale applications.”
Text2IMU: Advancing Human Activity Recognition by Text-Driven IMU Data Synthesis
Haeusler, L. O., Uhlenberg, L., & Amft, O. (2025) · Wearable and Implantable Body Sensor Networks · doi:10.1109/bsn66969.2025.11337798 · Cited by 5
“We evaluated our Text2IMU synthesis approach with three frequently used public benchmark datasets: RealWorld [16], PAMAP2 [17], and WISDM [18].”
RITA: Group Attention is All You Need for Timeseries Analytics
Liang, J., Cao, L., Madden, S. R., Ives, Z., & Li, G. (2024) · Proceedings of the ACM on Management of Data · doi:10.1145/3639317 · Cited by 5
“…multivariate timeseries dataset generated • HHAR dataset [46] contains sensing data of accelerometer collected • RWHAR RealWorld HAR dataset [48] covers 15 subjects performing • ECG dataset [34] consists of 10,000 EEG recordings for arrhythmia • MGH [6] is a EEG dataset collected by Mass. •…”
MethodTowards Understanding the Impact of Participant and its Wearable Devices in Federated Learning
Mishra, R., & Gupta, H. P. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3530818 · Cited by 5
“Similar to HAR, the REALWORLD [14] dataset involved 15 participants in collecting the sensory dataset of eight activities.”
Kostikis, N., Rigas, G., Tachos, N. S., Konitsiotis, S., & Fotiadis, D. I. (2020) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc44109.2020.9175913 · Cited by 5
“There is a growing interest in similar methods and the majority of them use machine learning algorithms to detect the actual placement in different activities [3-6].”
MethodRefined LSTM Network for Sensor-based Human Activity Recognition in Real World Scenario
Mekruksavanich, S., Jantawong, P., Hnoohom, N., & Jitpattanakul, A. (2022) · IEEE International Conference on Software Engineering and Service Science (ICSESS) · doi:10.1109/icsess54813.2022.9930218 · Cited by 5
“The REALWORLD16 dataset [22] is a publicly bench-marked HAR dataset containing sensor data based on acceleration, GPS, gyroscope, light, magnetic field, and sound level accumulated at a sampling frequency of 50 Hz for the following methods: descending and ascending stairs, jumping, lying, standing,…”
MethodWearable sensors for human activity recognition based on a self-attention CNN-BiLSTM model
Huafeng, G., Xiang, C., & Shi-qiang, C. (2023) · Sensor Review · doi:10.1108/sr-10-2022-0398 · Cited by 5
Model and Empirical Study on Multi-tasking Learning for Human Fall Detection
Nguyen, D., Pham, C., Argent, R., Caulfield, B., & Le‐Khac, N. (2024) · Vietnam Journal of Computer Science · doi:10.1142/s2196888824500040 · Cited by 5
Scheurer, S., Tedesco, S. A., Brown, K. N., & O’Flynn, B. (2019) · International Workshop on Sensor-based Activity Recognition and Interaction (iWOAR) · doi:10.1145/3361684.3361689 · Cited by 5
“[13] FUSION 7 10 90 ± 0 50 [1] MHEALTH 11 10 38 ± 0 50 [5] OPPORT 4 4 590 ± 258 30 [11] PAMAP2 12 9 81 ± 8 100 [15] REALWORLD 8 15 318 ± 42 50 [12] SAFESENS 17 11 91 ± 13 33 [9] SIMFALL 16 17 128 ± 8 25 [14] UTSMOKE 7 11 859 ± 7 50”
SensiX: A Platform for Collaborative Machine Learning on the EdgePreprint
Min, C., Mathur, A., Montanari, A., Acer, U. G., & Kawsar, F. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.06035 · Cited by 5
“For the additional details, refer to [18], [47] for the architecture of sensing models and [40], [46] for the datasets.”
Explainable Deep Learning Framework for Human Activity RecognitionPreprint
Huang, Y., Zhou, Y., Zhao, H., Riedel, T., & Beigl, M. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2408.11552 · Cited by 5
“To test OptiHAR in various scenarios and to keep consistency of the experiments with other works, we employ five widely used benchmark datasets in HAR, namely, DSADS [1], HAPT [14], OPPO [2], PAMAP2 [13], and RW [19].”
MethodFriemert, D., Schnur, D., Runkel, S., Borsch, J., Karamanidis, K., Dellen, B. K., Thieme, L., Fiedler, A., Jaekel, U., & Hartmann, U. (2025) · medRxiv · doi:10.1101/2025.05.29.25328474 · Cited by 5
CNN-based continuous authentication for digital therapeutics using variational autoencoder
Wang, C., Zhang, Y., Ma, Y., Chen, P., & Xiang, Y. (2024) · The Journal of Supercomputing · doi:10.1007/s11227-024-06490-2 · Cited by 4
“6, we can see that EE classifier achieves the lowest EER of 1.06 % and the highest accuracy of 98.96 % on gait dataset [31] (for the 7 th user), and it achieves the lowest EER of 2.33 % and the highest accuracy of 97.68 % in activity recognition dataset [32] (for the 10 th user).”
InfluentialWang, H., Liu, F., Li, X., Huang, H., Li, Y., & Sun, F. (2025) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3714394.3756209 · Cited by 4
“…human activity recognition datasets with varying sensor configurations, subjects, sampling frequencies, and activity categories, including PAMAP2 Dataset [11], UCI HAR Dataset [1], mHealth Dataset [3], RealWorld Dataset [13], KU-HAR Dataset [12], Kyutech IMU Dataset [2], and SHL Dataset [6, 15].”
Human Activity Recognition from Accelerometry, Based on a Radius of Curvature Feature
Cavita-Huerta, E., Reyes‐Reyes, J., Ugalde, H. M. R., Osorio-Gordillo, G. L., Escobar-Jiménez, R. F., & Alvarado, V. M. (2024) · Mathematical and Computational Applications · doi:10.3390/mca29050080 · Cited by 4
Exploring Cutout and Mixup for Robust Human Activity Recognition on Sensor and Skeleton Data
Dingeto, H., & Kim, J. (2024) · Applied Sciences · doi:10.3390/app142210286 · Cited by 4
Comparing LIME and SHAP Global Explanations for Human Activity Recognition
Alves, P., Delgado, J., Gonzalez, L., Rocha, A. R., Boccato, L., & Borin, E. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-79035-5_12 · Cited by 4
Semi-supervised Human Activity Recognition with individual difference alignment
Yang, Z., Li, T., Xu, Z., Huang, Z., Cao, Y., Li, K., & Ma, J. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.126976 · Cited by 4
Federated Learning with Heterogeneous Feature Adaptation for Human Activity Recognition
Wang, J., Schlagenhauf, T., & Maghsudi, S. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889340 · Cited by 4
“We evaluate our method using five HAR datasets, namely, UCI-HAR [20], HHAR [21], RealWorld [22], USC-HAD [23], and WISDM [24].”
Smartphone-based context-aware system for human activity recognition: dynamic and static methods
Attoumi, H., Achroufene, A., Saifi, R., Souici, L., & Boukredera, D. (2025) · Multimedia Tools and Applications · doi:10.1007/s11042-025-21053-0 · Cited by 4
“The datasets used to evaluate the two proposed methods are DOMINO [ 3 ], ExtraSensory [46] and RealWorld [47].”
Syed, A. S., Syed, Z. S., & Khalil, A. (2020) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2020.0111074 · Cited by 4
“The authors in [27] make use of accelerometer sensor data in the dataset in [28] and CNNs to differentiate between activities in the industry.”
MethodEffect of Combinations of Sensor Positions on Wearable-sensor-based Human Activity Recognition
Duan, Y., & Fujinami, K. (2023) · Sensors and Materials · doi:10.18494/sam4421 · Cited by 4
Is there an Optimal Technology to Provide Personal Supportive Feedback in Prevention of Obesity?
Sandri, S., Zheng, H., Engel, F., Moorhead, A., Wang, H., Bond, R., McTear, M., Molinari, A., Bouquet, P., & Hemmje, M. (2019) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm47256.2019.8983405 · Cited by 4
“As first results, the Real World Data [3] dataset will be used, for an initial requirement analysis and prototypical development.”
MethodXia, C., & Sugiura, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3134262 · Cited by 4
“As Figure 10 shows, the waist is given the highest ranking, which is consistent with the work of [20].”
BackgroundResultInfluential
Metadata
- Authors
- Sztyler, T., & Stuckenschmidt, H.
- Affiliations
- University of Mannheim (DE)
- Venue
- 2016 IEEE International Conference on Pervasive Computing and Communications (PerCom)
- Publisher
- IEEE
- Published
- 2016-03
- Pages
- 1-9
- ISSN
- 2474-249X (ISSN-L)
- OpenAlex
- W2340025709
- Cited by
- 369
- References
- 19
- Citations by year
- 2026: 30 · 2025: 55 · 2024: 60 · 2023: 38 · 2022: 39 · 2021: 35 · 2020: 41 · 2019: 33 · 2018: 22 · 2017: 11 · 2016: 5
- FWCI
- 8.21
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Pervasive Computing and Communications (publishedVersion)
- MADOC (University of Mannheim) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Indoor and Outdoor Localization Technologies
- Social Sciences › Transportation › Human Mobility and Location-Based Analysis
- Keywords
- wearable devices, activity recognition, accelerometer data, random forest classification, mobile device sensors, real-world data, ubiquitous computing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.