Deep Learning for Sensor-based Human Activity Recognition
The DOI matches its Crossref record; 151 of the 151 references it cites were found in a registry or an index.
Abstract
The vast proliferation of sensor devices and Internet of Things enables the applications of sensor-based activity recognition. However, there exist substantial challenges that could influence the performance of the recognition system in practical scenarios. Recently, as deep learning has demonstrated its effectiveness in many areas, plenty of deep methods have been investigated to address the challenges in activity recognition. In this study, we present a survey of the state-of-the-art deep learning methods for sensor-based human activity recognition. We first introduce the multi-modality of the sensory data and provide information for public datasets that can be used for evaluation in different challenge tasks. We then propose a new taxonomy to structure the deep methods by challenges. Challenges and challenge-related deep methods are summarized and analyzed to form an overview of the current research progress. At the end of this work, we discuss the open issues and provide some insights for future directions.
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
“This renders activity labeling difficult, reducing the number of available labels [8,23,30].”
From Spatial-Temporal Masked Autoencoder for Multi-Device Wearable Human Activity Recognition (2023) · cites this paper as [8]
Citations
Cites
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
Pan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387 · cited 1×
“Before taking a closer look at the factors that cause distribution discrepancy in sensory data, we briefly introduce transfer learning [102].”
BackgroundGreff, K., Srivastava, R. K., Koutnik, J., Steunebrink, B. R., & Schmidhuber, J. (2017) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2016.2582924 · Cited by 6,955 · cited 19×
“A multi-label RNN was created with each RNN cell responding to activity recognition of one occupant.”
BackgroundMethodInfluentialProceedings of the 19th International Joint Conference on Artificial Intelligence
Pujol, J. M., Delgado, J., Ramon Sangüesa, A., & Flache, A. (2005) · International Joint Conference on Artificial Intelligence · Cited by 5,771
Combining labeled and unlabeled data with co-training
Blum, A., & Mitchell, T. (1998) · Proceedings of the eleventh annual conference on Computational learning theory · doi:10.1145/279943.279962 · Cited by 5,495
Recurrent neural network based language model
Mikolov, T., Karafiát, M., Burget, L., Černocký, J., & Khudanpur, S. (2010) · Interspeech 2010 · doi:10.21437/interspeech.2010-343 · Cited by 5,390
Activity 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
Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761 · cited 1×
“eep hybrid model to explore different views of temporal dynamics is another attractive trend in the human activity recognition community. In light of the advantages of CNN and RNN, Ordóñez and Roggen [106] proposed to combine CNNs and LSTMs for both local and global temporal feature extraction. Xu et al. [162] adopted…”
A 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×
“data facilitates recognition and which part deteriorates that. 1.2 Deep Learning in Human Activity Recognition. Numerous previous works adopted machine learning methods in human activity recognition [79]. They highly rely on feature extraction techniques including time-frequency transformation [62], statistical approa…”
MethodInfluential
Research works printed but not linked: 37
- Alsheikh Mohammad Abu, 2016, Proceedings of the Workshops at the 30th AAAI Conference on Artificial Intelligence.
- Bengio Yoshua, 2012, Proceedings of ICML Workshop on Unsupervised and Transfer Learning. 17–36
- Brophy Eoin, 2018, Zhengwei Wang, Alan F. Smeaton, and Tomas E. Ward.
- 10.5555/3367032.3367223 · doi:10.5555/3367032.3367223
- Chen Ling, 2020, METIER: A deep multi-task learning based activity and user recognition model using wearable sensors. Proc. ACM Interact., Mob, Wear. Ubiq. Technol., 4, 1
- Dong Mingtao, 2018, Proceedings of the International Conference on Signal and Information Processing, Networking and Computers. Springer, 32–40
- Goodfellow Ian, 2014, Proceedings of the International Conference on Advances in Neural Information Processing Systems. 2672–2680
- Hammerla Nils Yannick, 2015, Proceedings of the 29th AAAI Conference on Artificial Intelligence.
- Hammerla Nils Y., 2016, Proceedings of the 25th International Joint Conference on Artificial Intelligence. 1533–1540
- Karpathy Andrej, 2016, Proceedings of the 4th International Conference on Learning Representations Workshop.
- Al Hafiz Khan Md Abdullah, 2018, Proceedings of the International Conference on Pervasive Computing and Communications. IEEE, 1–9.
- 10.5555/3164150.3164213 · doi:10.5555/3164150.3164213
- Krizhevsky Alex, Proceedings of the International Conference on Advances in Neural Information Processing Systems. 1097–1105
- Nicholas, Proceedings of the 16th International Workshop on Mobile Computing Systems and Applications. ACM, 117–122
- Lee Song-Mi, 2017, Proceedings of the IEEE International Conference on Big Data and Smart Computing (BigComp’17)
- Li Fei, 2011, Proceedings of the Workshops at the 25th AAAI Conference on Artificial Intelligence.
- Li Xinyu, Proceedings of the 8th Wireless of the Students, by the Students, and for the Students Workshop (S3@MobiCom’16)
- Li Xinyu, Proceedings of the 14th ACM Conference on Embedded Network Sensor Systems CD-ROM. ACM, 164–175
- Li Xinyu, 2017, Burd
- Ordóñez Morales Francisco Javier, 2016, Proceedings of the ACM International Symposium on Wearable Computers. ACM, 92–99
- Vishvak, Proceedings of the ACM International Symposium on Wearable Computers. ACM, 100–103
- Phan NhatHai, 2016, Proceedings of the 30th AAAI Conference on Artificial Intelligence.
- Pouyanfar Samira, 2018, A survey on deep learning: Algorithms, techniques, and applications, Comput. Surv., 51, 92
- Press Ofir, 2017, Language generation with recurrent generative adversarial networks without pre-training. arXiv preprint arXiv:1706.01399 · arXiv
- Qian Hangwei, 2018, Proceedings of the AAAI Conference on Artificial Intelligence.
- Saeedi Ramyar, Proceedings of the IEEE International Conference on Big Data. IEEE, 473–479
- Serrano Sofia, Proceedings of the 57th Conference of the Association for Computational Linguistics (ACL’19)
- van Kasteren Tim L. M., 2011, Kröse
- Xue Li, 2018, Understanding and improving deep neural network for activity recognition. arXiv preprint arXiv:1805.07020 · arXiv
- 10.5555/2832747.2832806 · doi:10.5555/2832747.2832806
- Zebin Tahmina, Proceedings of the IEEE Conference on Sensors (SENSORS’16)
- Zhang Dalin, 2018, Proceedings of the AAAI Conference on Artificial Intelligence, 32
- Zhang Mi, Proceedings of the ACM Conference on Ubiquitous Computing. ACM, 1036–1043
- 10.5555/3304889.3305093 · doi:10.5555/3304889.3305093
- Zhang Yanyi, Proceedings of the 16th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN’17)
- Zhang Yong, 2018, Human activity recognition based on time series analysis using U-Net. arXiv preprint arXiv:1809.08113 · arXiv
- Zheng Yi, Proceedings of the International Conference on Web-age Information Management. Springer, 298–310
Cited by
Items 201–300 of 503
Wang, J., Al‐qaness, M. A. A., Ni, S., & Tang, C. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3540882 · Cited by 6
“1) Traditional Sensor-Based Approaches: Early methods of human activity recognition relied heavily on sensor-based approaches [12], [13], where wearable sensors or fixed sensors were used to collect data about human movements and activities.”
Differentiable Prior-Driven Data Augmentation for Sensor-Based Human Activity Recognition
Zhang, Y., Gao, Q. H., Hu, R., Ding, Q., Li, B., & Guo, Y. (2025) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2025.3565414 · Cited by 6
“Then, the normalized signals are segmented into smaller patches using a sliding window [15], with specific segmentation window sizes provided in Table I.”
Client-Unbiased Skeletal Action Recognizer in Federated Learning
Zhu, X., Shu, X., & Tang, J. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3586511 · Cited by 8
“Yet, these devices often capture sensitive personal information [6], [7], and traditional RGB-based HAR methods [8] can inadvertently reveal identities through appearance cues [9], attire, and behavior patterns [10].”
Optimizing Accuracy-Efficiency Trade-Offs of On-Device Activity Inference With Star Operation
Chen, G., Fu, Z., Sha, Y., Xiong, D., Zhang, L., Wu, H., & Song, A. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3619549 · Cited by 6
“…sensor-based human activity recognition (HAR) into a hot research topic in ubiquitous computing community, spanning a wide range of practical applications in various real-world scenarios like health management [1], [2], activity tracking, gesture recognition, and human-computer interaction [3].”
Yan, S., Lin, K.-J., & Wang, H. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3386794 · Cited by 7
“Ambient sensors can be embedded in an environment to capture the interactions between humans and their environment, and to detect multi-occupant activities or in-door localizing [3].”
BackgroundInfluentialYao, M., Zhang, L., Cheng, D., Liu, J., Wang, L., Wu, H., & Song, A. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3412395 · Cited by 7
“Among them, the fundamental data preprocessing pipeline usually includes handling missing values, noise filtering, data normalization, and data segmentation, which have been typically implemented in various HAR applications, as summarized in previous survey literature [1] and [6].”
BackgroundMethodA Spatio-temporal Graph Transformer driven model for recognizing fine-grained data human activity
Yan, M., Zhang, G., & Ye, C. (2024) · Alexandria Engineering Journal · doi:10.1016/j.aej.2024.06.042 · Cited by 7
Dridi, J., Amayri, M., & Bouguila, N. (2024) · International Conference on Intelligent Information Technology (ICIIT) · doi:10.1145/3654522.3654541 · Cited by 7
Towards Enhanced Human Activity Recognition for Real-World Human-Robot Collaboration
Yalçinkaya, B., Couceiro, M. S., Pina, L., Soares, S., Valente, A. L. G., & Fabio Remondino, A. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610664 · Cited by 8
“The recent deep learning popularity, with Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), has significantly improved the accuracy and robustness of HAR systems [13].”
BackgroundWang, 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
“With advances in neural networks, deep learning-based models have dominated the HAR domain in the last decade [9], [10].”
BackgroundDevice-Free Human Activity Recognition: A Systematic Literature Review
Moghaddam, M. G., Shirehjini, A. A. N., & Shirmohammadi, S. (2024) · IEEE Open Journal of Instrumentation and Measurement · doi:10.1109/ojim.2024.3502885 · Cited by 13
Cross-User Activity Recognition Using Deep Domain Adaptation With Temporal Dependency Information
Ye, X., & Wang, K. I. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3552453 · Cited by 9
“However, these methods largely rely on the assumption that the training and testing data are derived from the same distribution, thus ensuring the model’s generalizability, that is, all data are i.i.d. [15].”
Time-Series Contrastive Learning Against False Negatives and Class Imbalance
Jin, X., Wang, J., Ou, X., Liu, L., & Lin, Y. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3568387 · Cited by 14
“Another example is physiological time-series scenarios, like electroencephalogram (EEG)-based sleep analysis [9] and human activity recognition [10], where it is even necessary to involve multiple experts working collaboratively to ensure precise and reliable annotations.”
Static and Dynamic Human Activity Detection Using Multi CNN-ELM Approach
Ankalaki, S., & Thippeswamy, M. N. (2021) · Lecture notes in electrical engineering · doi:10.1007/978-981-16-1338-8_18 · Cited by 6
VALERIAN: Invariant Feature Learning for IMU Sensor-based Human Activity Recognition in the Wild
Hao, Y., Wang, B., & Zheng, R. (2023) · IEEE/ACM International Conference on Internet-of-Things Design and Implementation (IoTDI) · doi:10.1145/3576842.3582390 · Cited by 8
“Finally, we believe significant efforts should be made to build in-the-wild datasets and benchmarks for IMU-based HAR.”
BackgroundInfluentialLeveraging a Smartwatch for Activity Recognition in Salat
Eliza, I. J., Al-Nabhan, N., Noor, J., Rahaman, M., & Islam, A. B. M. A. A. (2023) · IEEE Access · doi:10.1109/access.2023.3311261 · Cited by 9
“Recently, Deep Learning (DL) methods such as recurrent neural networks (RNN), LSTMs, autoencoders, and their variations have been proven to provide state-of-the-art results on challenging activity recognition tasks with little or no data feature engineering [137].”
BackgroundMulti-DGI: Multi-head Pooling Deep Graph Infomax for Human Activity Recognition
Chen, Y., Zhu, H., & Chen, Z. (2024) · Mobile Networks and Applications · doi:10.1007/s11036-024-02306-y · Cited by 9
Benis, A., Haghi, M., Tamburis, O., Darmoni, S. J., Julien Grosjean, A., & Deserno, T. M. (2023) · Yearbook of Medical Informatics · doi:10.1055/s-0043-1768742 · Cited by 6
“Behavioural, such as gait analysis and human activity recognition [45]; 3.”
BackgroundYe, N., Zhang, L., Xiong, D., Wu, H., & Song, A. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3555985 · Cited by 7
“These sensors, known for their compact size, high accuracy, and low cost, can be seamlessly integrated into portable devices such as smartphones and bracelets to detect and classify daily human activities [1], [2], [3], [4].”
Optimizing Human Activity Recognition Using Stacked CNN with Ensemble Learning
Yadav, D., Rani, D., & Verma, O. P. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.04.391 · Cited by 7
SAGoG: Similarity-Aware Graph of Graphs Neural Networks for Multivariate Time Series Classification
Wang, S., Zhang, Y., Lin, X., Hu, Y., Huang, Q., & Yin, B. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3572216 · Cited by 6
“M ULTIVARIATE Time Series (MTS) are widely available in many fields such as action recognition [1], health-care [2], traffic management [3] and energy management [4], providing indispensable support for data analysis and decision-making in real-world scenarios.”
Agent-in-the-loop to distill expert knowledge into artificial intelligence models: a survey
Gao, J., Zhang, Y., Chen, Y., Dong, Y., Chen, Y., Song, S., Tang, B., & Gu, Y. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11255-1 · Cited by 10
Wei, Q., Huang, J., Gao, Y. C., & Dong, W. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3749509 · Cited by 21
Liu, M., Geißler, D., Bian, S., Zhou, B., & Lukowicz, P. (2025) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops65533.2025.00083 · Cited by 9
MicroFluID: A Multi-Chip RFID Tag for Interaction Sensing Based on Microfluidic Switches
Sun, W., Chen, Y., Chen, Y., Zhang, X., Zhan, S., Li, Y., Wu, J., Han, T., Mi, H., Wang, J., et al. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3550296 · Cited by 12
“As numerous previous work that adopted machine learning methods are normally restricted to heuristic feature extractions [5], recent researches tend to apply deep learning to mitigate the challenges of environment independent and user invariant gesture sensing via modeling variant features [7, 27,…”
BackgroundDeep Incremental Learning for Personalized Human Activity Recognition on Edge Devices
Younan, S., & Abu-Elkheir, M. (2022) · Canadian Journal of Electrical and Computer Engineering · doi:10.1109/icjece.2022.3199227 · Cited by 7
“A comprehensive body of research work explored the development of machine learning and deep learning algorithms and architectures for HAR [6], [7], [8], with the main focus being on using sensors to provide data for activity recognition and analysis.”
BackgroundHuman Activities Recognition using Semi-Supervised SVM and Hidden Markov Models
García, S. M., Baena, C. A. H., & Salcedo, A. F. C. (2022) · TecnoLógicas · doi:10.22430/22565337.2474 · Cited by 7
“These methods have encompassed the identification of human physical activity and have achieved reliable results [16]−[18].”
MethodDjemili, R., & Zamouche, M. (2023) · International Journal of Computers and Applications · doi:10.1080/1206212x.2023.2198785 · Cited by 5
“However, bulky wearable sensors are inescapably becoming obsolete due to the ever-growing popularity of smaller, more compact, and immensely more powerful, and advanced electronic devices such as smartphones or smartwatches [13,16,17].”
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
“A notable example is the work by [30], who applied attention-based models to HAR, demonstrating improved performance compared to traditional CNN and LSTM approaches.”
BackgroundMethodTakeda, N., Legaspi, R., Nishimura, Y., Ikeda, K., Plötz, T., & Chernova, S. (2024) · IEEE International Conference on Big Data Analytics (ICBDA) · doi:10.1109/icbda61153.2024.10607364 · Cited by 7
“Numerous research efforts have been focused on HAR within households to monitor residents’ well-being and to support elderly people in their daily activities [14], [15].”
BackgroundHan, C., & Liu, P. (2024) · IEEE Access · doi:10.1109/access.2024.3443339 · Cited by 6
“A cutting-edge deep learning method for recognising human motion activity was presented by Chen et al. [14].”
MethodA Virtual-Label-Based Hierarchical Domain Adaptation Method for Time-Series Classification
Yang, W., Cheng, L., Ragab, M., Wu, M., Pan, S. J., & Chen, Z. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3445879 · Cited by 5
Fan, Y., Jitpattanakul, A., & Mekruksavanich, S. (2025) · International Conference on Big Data Analytics and Practices (IBDAP) · doi:10.1109/ibdap65587.2025.11145849 · Cited by 5
“…automatically from sensor data holds transformative potential across various applications, including medical monitoring [1]– [4], personal fitness assessment [5], [6], fall prevention systems for older adults [7]–[11], physical rehabilitation [12], and intelligent home environments [13]–[16].”
Transfer Learning Model Knowledge Across Multi-Sensors Locations Over Body Sensor Network
Qian, X., Chen, H., Cai, Y. M., Chu, K., Xu, W., & Huang, M.-C. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3166187 · Cited by 6
“Due to the ease of collection and set up [25], as well as the protection of privacy [26], the sensor-based HAR approaches have become more prevalent and are widely used all over the world [27].”
BackgroundDigital methods for sensory media research: Toolmaking as a critical technical practice
Chao, J., van Geenen, D., Gerlitz, C., & van der Vlist, F. (2024) · Convergence The International Journal of Research into New Media Technologies · doi:10.1177/13548565241226791 · Cited by 8
“(e .g. Chen et al., 2021).”
Wang, L., Yan, Y., Huang, D., Pan, Y., Cheang, C. F., Luo, K., & Li, J. (2025) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2025.3540853 · Cited by 6
“With the development of deep learning, HAR models based on deep learning techniques are widely applied in the feature engineering and recognition phases of HAR tasks to efficiently extract data features and recognize activities [9], [10], [11], [12].”
Ayub, M., & El-Alfy, E.-S. M. (2025) · IEEE Access · doi:10.1109/access.2025.3631392 · Cited by 6
“Various deep learning approaches have been explored for HAR [22] and more specifically for sensor-based HAR [23].”
mm-ARnet: Exploring Millimeter Wave Radar Point Clouds for Human Action Recognition
Hao, Z., Xiao, J., Wang, Y., Wang, G., & Li, F. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3649493 · Cited by 7
“Wearable devices [1] require users to carry or wear hardware, limiting practicality due to charging needs and placement constraints.”
Kianpisheh, M., Mariakakis, A. T., & Truong, K. N. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3643500 · Cited by 8
Deep Learning for Frailty Classification Using IMU Sensor Data: Insights From FRAILPOL Database
Amjad, A., Szczęsna, A., Błaszczyszyn, M., Sacha, J., Sacha, M., Feusette, P., Wolański, W., Konieczny, M., Borysiuk, Z., & Khan, B. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3510626 · Cited by 4
Lee, H.-J., & Buu, S.-J. (2025) · IEEE Access · doi:10.1109/access.2025.3568505 · Cited by 7
“In recent years, there has been a growing interest in non-invasive and contactless sensing technologies, with Wi-Fi signal-based human activity analysis emerging as a critical research area for the development of smart environments [1], [2], [3].”
Combining Clustering Undersample and Ensemble Learning for Wearable Fall Detection
Zhang, M., & Gong, D. (2022) · IEEE International Conference on Real-time Computing and Robotics (RCAR) · doi:10.1109/rcar54675.2022.9872285 · Cited by 4
“Falling activities are regarded as positive samples, and daily activities are regarded as negative samples[11].”
BackgroundMAD-MEL: Combining Entity and Metric Learning for Anomaly Detection in Multivariate Time Series
Kumari, J. J. J., Mathew, J., & Mondal, A. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3339335 · Cited by 4
“In general, time series data is analyzed for three broad categories of tasks: classification [1], prediction [2], and anomaly detection [3].”
SemNet: Learning semantic attributes for human activity recognition with deep belief networks
Venkatachalam, S., Nair, H., Zeng, M., Tan, C. S., Mengshoel, O. J., & Shen, J. P. (2022) · Frontiers in Big Data · doi:10.3389/fdata.2022.879389 · Cited by 4
“Further, many deep learning approaches have been applied in activity recognition task (Plötz et al., 2011; Zeng et al., 2014a, 2017, 2018; Chen et al., 2021).”
BackgroundThakur, N., Singh, S., Gupta, A., Jain, K., Jain, R., Peraković, D., Nedjah, N., & Rafsanjani, M. K. (2022) · International Journal of Software Science and Computational Intelligence · doi:10.4018/ijssci.311445 · Cited by 5
Daily unbalanced action recognition based on active learning
Liu, Y., Li, Z., Huan, Z., Zhou, B., Shen, S., & Gao, S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16181-4 · Cited by 4
“HAR is mainly used in the fields of smart home system, health management, interactive experience and medical diagnosis [38], and its development is crucial to human beings [13].”
Self-citationJiang, L., & Lu, W. (2023) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2023.1275645 · Cited by 7
Ye, N., Zhang, L., Xiong, D., Wu, H., & Song, A. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3458441 · Cited by 5
“Chen et al. [3] and Hammerla et al. [28] have provided a detailed guideline for researchers who seek to adopt deep neural networks in HAR problem setting.”
Human activity recognition for static and dynamic activity using convolutional neural network
Minarno, A. E., Kusuma, W. A., & Kurniawan, Y. A. (2021) · TELKOMNIKA (Telecommunication Computing Electronics and Control) · doi:10.12928/telkomnika.v19i6.20994 · Cited by 4
Multitemporal Sampling Module for Real-Time Human Activity Recognition
Park, J., Lim, W.-S., Kim, D., & Lee, J. (2022) · IEEE Access · doi:10.1109/access.2022.3176606 · Cited by 5
“In HAR literature, many comprehensive surveys have emphasized the importance of real-time applications [1], [6], [8], [27], [28].”
BackgroundMethodInfluentialHuman Activity Recognition Based on Hybrid Learning Algorithm for Wearable Sensor DataPreprint
Athota, R. K., & Sumathi, D. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4162745 · Cited by 4
Hash-MMDC: Enhancing human activity recognition with hash-based optimization
Cheng, S., Liu, Q., & Wan, J. (2025) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-025-00297-8 · Cited by 3
“For instance, CNNs often struggle to capture long-range temporal dependencies, and RNNs can suffer from vanishing gradients on long sequences (Chen et al. 2021).”
Li, Z., Fu, M., Fu, B., & Liu, Y. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-04888-w · Cited by 4
“However, several challenges remain: most HAR models still use predetermined observation lengths without evaluating sufficiency [5], often ignore inter-sensor relationships in multimodal data [6], and increasing model complexity can degrade efficiency on wearable devices [7].”
Self-citationOishi, N., Birch, P. M., Roggen, D., & Lago, P. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3661047 · Cited by 9
“To address this, data augmentation is often employed [9].”
Bian, S., Liu, M., Ray, L. S. S., Zhou, B., Guo, B., Yu, Z., Plötz, T., Lukowicz, P., Yuan, S., & Rey, V. F. (2026) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3810230 · Cited by 16
Self-citationMultiuser Behavior Recognition Module Based on DC-DMN
An, J., Cheng, Y.-S., He, X., Gui, X., Wu, S., & Zhang, X. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3133870 · Cited by 3
Deep Neural Network-Based Business Data Classification in Intelligent Business Management
Wang, B.-H. (2022) · Mathematical Problems in Engineering · doi:10.1155/2022/7104750 · Cited by 3
“When the construction of IT infrastructure system is maturing, the establishment and implementation of BI system will provide a unified view for the enterprise, which means that the application of BI can centralize and present the data from various systems, provide the management with various useful data needed, and i…”
BackgroundRetractedArokiaraj, S. P., & Viswanathan, N. (2023) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-221551 · Cited by 6
Less is more: Efficient behavioral context recognition using Dissimilarity-Based Query Strategy
Akram, A., Farhan, A. A., & Basharat, A. (2023) · PLoS ONE · doi:10.1371/journal.pone.0286919 · Cited by 3
Emovo: A real-time anger detector on the smartphone using acoustic signal
Tong, Y., Mo, W., & Sun, Y. (2023) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3594806.3594833 · Cited by 3
“Computation and analysis of this data can be accomplished through deep learning techniques[3].”
BackgroundSample-Based Human Movement Detection for Interactive Videos Applied to Performing Arts
Rodrigues, R., Diogo, J., Jürgens, S., et al. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-42286-7_32 · Cited by 3
Rosita, E., Utomo, A. P., Azizah, S. A., & Sukoco, S. (2023) · Jurnal Biologi · doi:10.47134/biology.v1i3.1973 · Cited by 17
Accurate Classification of Sport Activities Under Tiny Deployability Constraints
Mazinani, A., Davoli, L., Pau, D. P., & Ferrari, G. (2023) · IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS) · doi:10.1109/iotais60147.2023.10346056 · Cited by 3
“Hence, the diversity and complexity of daily activities, as well the necessity to ensure the accuracy of the sensors and the produced data are a few of the difficulties involved with HAR [5,6].”
BackgroundNanda, S., & Dutta, S. K. (2023) · IEEE Silchar Subsection Conference (SILCON) · doi:10.1109/silcon59133.2023.10404163 · Cited by 3
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
“Thanks to the proliferation of smart devices and the IoT, sensors can be seamlessly integrated into portable devices like phones and watches, as well as non-portable objects such as cars, walls, and furniture [24], [25].”
BackgroundDeep similarity segmentation model for sensor-based activity recognition
Baraka, A. M., & Noor, M. H. M. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18933-2 · Cited by 4
“Since the accelerometer and gyroscope are primarily designed to capture human movements [33] and are not able to capture speech, both activities are treated in this study as belonging to the same class of activity [29].”
Feng, H., Shen, Q., Song, R., et al. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4460033 · Cited by 3
Machine Learning for Human Motion Change DetectionPreprint
Lin, J.-J., Hsu, C.-K., Hsu, W., et al. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4474404 · Cited by 3
ContextGPT: Infusing LLMs Knowledge into Neuro-Symbolic Activity Recognition Models
Arrotta, L., Bettini, C., Civitarese, G., & Fiori, M. (2024) · arXiv · doi:10.48550/arxiv.2403.06586 · Cited by 3
“The analysis of sensor data gathered from mobile and wearable devices for Human Activity Recognition (HAR) has been extensively researched [1], [2].”
BackgroundA Survey of Motion Data Processing and Classification Techniques Based on Wearable Sensors
Xiaoqiong, X., Xuemei, 熊. X., Keda, Z., & Chao, L. (2023) · IgMin Research · doi:10.61927/igmin123 · Cited by 2
GNet-FHO: A Light Weight Deep Neural Network for Monitoring Human Health and Activities
Athota, R. K., & Sumathi, D. (2024) · IEEE Access · doi:10.1109/access.2024.3426075 · Cited by 4
“Recent studies in HAR have primarily concentrated on utilizing Machine Learning [8] and Deep Learning [9] techniques.”
MethodZhang, J., Peng, C., Liu, Z., Feng, L., Wu, Y., Dong, Y., & Xu, D. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3572220 · Cited by 6
“With the advancement of deep learning (DL) techniques [6], [7], [8], [9] in time series classification (TSC) for HAR tasks, the deployment of trained models directly on edge devices has become increasingly feasible [10], [11].”
Heravi, M. Y., Jang, Y., & PhD, H. C. (2025) · Journal of Construction Engineering and Management · doi:10.1061/jcemd4.coeng-15888 · Cited by 6
Barshan, B., & Koşar, E. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3577925 · Cited by 5
“Both the HAR and UIR problems involve classification tasks (of activities and users, respectively) for which extraction of discriminative features is essential [13].”
On Neural Inertial Classification Networks for Pedestrian Activity Recognition
Yampolsky, Z., Kruzel, O., Fekson, V. K., & Klein, I. (2025) · IEEE/ION Position Location and Navigation Symposium · doi:10.1109/plans61210.2025.11028462 · Cited by 5
SkeFi: Cross-Modal Knowledge Transfer for Wireless Skeleton-Based Action Recognition
Huang, S., Zhou, Y., & Yang, J. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3647603 · Cited by 2
“Scholars have delved into various sensing technologies to capture spatiotemporal and kinematic insights into human activities [31]–[33].”
LLaSA: A Sensor-Aware LLM for Natural Language Reasoning of Human Activity from IMU Data
Imran, S. A., Khan, M. N. H., Biswas, S., & Islam, B. (2025) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3714394.3756187 · Cited by 67
“Advances in Large Language Models (LLMs) and Visual Question Answering (VQA) have shown how multimodal systems can improve human interaction with complex data [14, 15, 17, 24].”
Chen, Y.-L., Chen, Y.-C., Lou, K., Chen, Y.-S., Li, H., & Jiang, Z.-D. (2025) · 2025 4th International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) · doi:10.1109/icicml67980.2025.11333691 · Cited by 1
“…adopt fixed-weight strategies, failing to consider modality advantages across different pickleball action phases (e.g., visual information is more important during preparation phase, while sensor data is more critical during hitting phase), resulting in low fusion efficiency (Chen et al., 2021)[1].”
InfluentialLee, D., Park, D., & Kang, Y.-S. (2026) · Journal of Computational Design and Engineering · doi:10.1093/jcde/qwag011 · Cited by 2
A multimodal spatio-temporal graph neural network framework for fatigue detection in tennis serving
Yuan, Y., Wang, L., Hui, J., & Jiao, J. (2026) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2026.1828589 · Cited by 4
“Multi-sensor fusion and wearable-sensor-based activity recognition have been widely reviewed and applied in human movement analysis (Aguileta et al., 2019; Taborri et al., 2016; Chen et al., 2012; Chen et al., 2021; Wu et al., 2021).”
Deep-3DConvNet: A Network to Detect Abnormal Activities at Megastores
Ansari, M. A., & Singh, D. K. (2022) · 2022 IEEE Bombay Section Signature Conference (IBSSC) · doi:10.1109/ibssc56953.2022.10037326 · Cited by 2
“Therefore, an automated and proactive video surveillance system [3, 4, 5] is in serious demand to seamlessly examine video feeds and produce an alert when anyone is found engaged in such kind of abnormal or unusual behavior at megastores.”
BackgroundModaldrop: Modality-Aware Regularization for Temporal-Spectral Fusion in Human Activity Recognition
Zeng, X., Chen, Y., Xu, B., & Zhang, T. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095880 · Cited by 3
“It supports extensive applications such as smart homes, healthcare and enhanced manufacturing, etc [1].”
BackgroundAutomatic cattle activity recognition on grazing systems
Agudelo, J. F. R., Mazo, S. B., Ochoa, S. L. P., & Noguera, J. R. R. (2022) · Biotecnología en el Sector Agropecuario y Agroindustrial · doi:10.18684/rbsaa.v20.n2.2022.1940 · Cited by 3
SRNF-PAR: Strawberry and Recurrent Neural Framework-based Physiotherapy Activity Recognition
Deotale, D., Verma, M., Suresh, P., & Bharadwaj, A. (2024) · International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · doi:10.1109/icrito61523.2024.10522121 · Cited by 2
“Estimating the pose is a technique used to examine the static [6] person's body position, providing useful information on postural rightness.”
MethodKasubi, J., Huchaiah, M. D., Gad, I., & Hooshmand, M. K. (2024) · Journal of Information Systems and Telecommunication (JIST) · doi:10.61186/jist.36294.12.46.127 · Cited by 2
Assessment of IMU Configurations for Human Activity Recognition Using Ensemble Learning
Murphy, S. J., & Vitali, R. V. (2024) · IEEE Access · doi:10.1109/access.2024.3440215 · Cited by 3
“For some deep learning applications, raw data from the sensors can be input directly into the algorithm that then determines what features to use [19], [20].”
BackgroundThe SES framework and Frequency domain information fusion strategy for Human activity recognition
Wu, H., Feng, H., Shi, L., Zhang, H., & Xu, H. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10650417 · Cited by 2
“Classic methods include adding random noise in the time dimension [16] and frequency distortion [17], slicing or cropping [18], scaling [16], random warping, etc.”
MethodWang, X., Zhang, L., Cheng, D., Tang, Y., Wang, S., Wu, H., & Song, A. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3481549 · Cited by 3
“As summarized in previous survey literature [1], [3], [9], the most widely employed segmentation strategy is sliding window.”
BackgroundMethodInfluentialHuman Activity Recognition (HAR) via Artificial Intelligence (AI) on the Edge
Alteneiji, A., Suliman, A., Sawalha, G., Poon, K., & AlDurra, T. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-5035-1_48 · Cited by 2
Dhage, J., & Gulve, A. K. (2025) · Engineering, Technology & Applied Science Research · doi:10.48084/etasr.10499 · Cited by 1
“Unlike traditional RNNs, LSTM networks are equipped with a memory cell that can maintain information over extended periods, making them particularly effective for tasks that require understanding temporal dynamics in human activities [8].”
Personal Data Protection in Smart Home Activity Monitoring for Digital Health: A Case Study
Bettini, C., Moradbeikie, A., & Civitarese, G. (2025) · 2025 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) · doi:10.1109/percomworkshops65533.2025.00108 · Cited by 1
Hu, S., Liu, J., Zhang, C., Zhu, X., & Li, L. (2024) · International Conference on Mobility, Sensing and Networking (MSN) · doi:10.1109/msn63567.2024.00018 · Cited by 4
“Environmental sensors collect data from microphones and vibration sensors to monitor and analyze activity-induced changes [11], [12].”
A uniform approach to HAR recognition in unobtrusive indoor monitoring systems
Salice, F., Mangano, D., Bouchard, K., & Comai, S. (2025) · Journal of Reliable Intelligent Environments · doi:10.1007/s40860-025-00255-1 · Cited by 2
Timely Classification of Hierarchical Classes
Abdelzaher, T., Baruah, S., Burns, A., & Hu, Y. (2025) · IEEE Real-Time Systems Symposium (RTSS) · doi:10.1109/rtss66672.2025.00025 · Cited by 3
CASL: Capturing Activity Semantics Through Location Information for Enhanced Activity Recognition
Zhang, X., Shan, C., Zhu, T., Chen, L., Zhou, F., & Ning, H. (2023) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2023.3238064 · Cited by 2
“Thelackoflabeledsensordataduetocostisknownasannotation scarcity [2].”
Samiee, N., Manouchehri, N., & Bouguila, N. (2023) · IEEE International Conference on Industrial Technology (ICIT) · doi:10.1109/icit58465.2023.10143109 · Cited by 2
“…has always been one of the most important and rapidly growing scientific fields, it has become increasingly common to use these techniques in the biomedical sector to identify latent patterns and, therefore, to extract valuable information from medical records [5], [6], [7], [8], [9], [10].”
BackgroundDDTM: A Distance-Based Data Transformation Method for Time Series Classification
Xu, H., Wang, K., Sun, W., Chen, M., Li, H., & Zhao, H. (2024) · Communications in computer and information science · doi:10.1007/978-981-99-9109-9_10 · Cited by 2
A Novel Human Activity Recognition Model
Zeng, X., Huang, M., Zhang, H., Ji, Z., & Ганчев, И. (2023) · International Conference on Mathematics and Computers in Sciences and Industry (MCSI) · doi:10.1109/mcsi60294.2023.00024 · Cited by 2
Sussex-Huawei Locomotion Recognition Using Machine Learning and Deep Learning with Multi-sensor data
Wang, H., Huang, H., Wang, J., & Sun, F. (2024) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3675094.3678457 · Cited by 3
Feature Extraction using Variational Auto-encoder for Radar-based Posture Detection Systems
Casmin, E., & Oliveira, R. (2024) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2024-spring62846.2024.10683040 · Cited by 2
Metadata
- Authors
- Chen, K., Zhang, D., Yao, L., Guo, B., Yu, Z., & Liu, Y.
- Affiliations
- Aalborg University (DK)
- UNSW Sydney (AU)
- Northwestern Polytechnical University (CN)
- Michigan State University (US)
- Venue
- ACM Computing Surveys
- Published
- 2021-05-24
- Volume
- 54
- Issue
- 4
- Pages
- 1-40
- ISSN
- 0360-0300 (ISSN-L)
- 1557-7341
- OpenAlex
- W3164845984
- Cited by
- 690
- References
- 144
- Citations by year
- 2026: 83 · 2025: 137 · 2024: 172 · 2023: 149 · 2022: 101 · 2021: 45
- FWCI
- 41.94
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- ACM Computing Surveys (publishedVersion, public-domain)
- VBN Forskningsportal (Aalborg Universitet) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Human Pose and Action Recognition
- Physical Sciences › Engineering › Biomedical Engineering › Gait Recognition and Analysis
- Keywords
- human activity recognition, sensor-based activity recognition, deep learning, Internet of Things, sensor fusion, activity recognition dataset, wearable sensors
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.