Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Open accessThe DOI matches its Crossref record; 48 of the 49 references it cites were found in a registry or an index; 8 of the 23 sources its claims cite were checked for support.
Abstract
Introduces DeepConvLSTM, a deep learning architecture combining convolutional and LSTM recurrent units for multimodal wearable activity recognition, outperforming non-recurrent deep networks and handcrafted feature baselines by 4% to 9%.
Overview
CoChat Summary
TL;DR
The paper proposes DeepConvLSTM, a deep framework combining four convolutional layers with two LSTM recurrent layers for wearable activity recognition from raw multimodal sensor data. It reports improved F1 scores over a baseline CNN and prior results on the OPPORTUNITY and Skoda datasets, including 0.915 F1 for OPPORTUNITY gestures with the Null class and 0.958 F1 on Skoda.
Key findings
- On OPPORTUNITY, DeepConvLSTM achieved F1 scores of 0.930 for modes of locomotion without Null, 0.895 for modes with Null, 0.866 for gestures without Null, and 0.915 for gestures with Null; the baseline CNN scored 0.912, 0.878, 0.783, and 0.883, respectively. The paper reports a 6% average improvement over the best OPPORTUNITY challenge submissions and more than 9% improvement on gesture recognition without Null.
- On Skoda, DeepConvLSTM reached 0.958 F1, compared with 0.884 for the baseline CNN, 0.893 for CNN [43], and 0.861 for CNN [23], improving the best reported result by 6%.
- Multimodal fusion improved gesture recognition on OPPORTUNITY without Null: 0.689 F1 with accelerometers only, 0.611 with gyroscopes only, 0.745 with accelerometers + gyroscopes, 0.839 with accelerometers + gyroscopes + magnetic sensors, and 0.864 with the full OPPORTUNITY sensor set.
- DeepConvLSTM had 999,122 parameters versus 7,445,458 for the baseline CNN, a 600% increase for the baseline; training took 340.3 min versus 282.2 min, and classification took 6.68 s versus 5.43 s, with DeepConvLSTM classifying almost 15 min of data per second and achieving 900× real-time on the GPU used.
Methods
- DeepConvLSTM uses four convolutional layers with 64 feature maps each, followed by two dense LSTM layers with 128 cells and a softmax output; the baseline CNN shares the convolutional layers but uses two non-recurrent dense layers with 128 units. Models were trained supervised with cross-entropy loss, mini-batch gradient descent with RMSProp, learning rate 10e-3, decay factor 0.9, mini-batches of 100 segments, and dropout p=0.5 on dense-layer inputs.
- Inputs were sliding windows of 500 ms with 250 ms step size, using raw sensor data with minimal preprocessing: linear interpolation for missing values and per-channel normalization to [0,1]. Evaluation used weighted F1 to address class imbalance.
- Experiments used the OPPORTUNITY dataset (on-body sensors, 113 channels, 30 Hz; Task A: 5-class modes of locomotion/postures; Task B: 18-class sporadic gestures) and the Skoda dataset (one subject, 10 right-arm 3D accelerometers, decimated to 30 Hz, 10 manipulative gestures, about 3 h and about 70 repetitions per gesture). Analyses included comparison to baseline CNN and published results, confusion matrices, sensor-modality fusion, and hyperparameter evaluations of sequence length and number of convolutional layers.
Limitations
- The paper states that DeepConvLSTM has not yet been implemented on a wearable device; the GPU used clearly outperforms current high-end wearable computational power, and training is best envisioned server-side. It also states that pooling was omitted because the sliding-window setup limits downsampling, though pooling could be useful without that requirement, and that training time increases with the number of layers, creating a performance–computational budget trade-off.
- Evident from the text: Skoda evaluation uses one subject and only right-arm sensors, and OPPORTUNITY testing is limited to ADL4 and ADL5 for Subjects 2 and 3, which constrains generalizability. The text does not report cross-validation or statistical significance testing.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures



![Figure 4 Placement of on-body sensors used in the OPPORTUNITYdataset (left: inertial measurements units; right: 3-axis accelerometers) [7].](https://figures.cochat.ai/paper-figures/27c0e4b39d104fe2b5315587d736fd2d/deep-convolutional-and-lstm-recurrent-neural-networks-for-multimodal-wearable-figure-4.jpg)




Verification
Claims the paper makes
5 Valid · 5 Not valid · 13 Not checked · 99 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“The choice of the number of recurrent layers is made following the results presented in [33], where the authors showed that a depth of at least two recurrent layers is beneficial when processing sequential data.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.1. DeepConvLSTM
What the sources say
- Supportedcites [33] Karpathy A. Johnson J. Li F.F. Visualizing and understanding recurrent networks Cornell Univ. Lab. 2015 arXiv:1506.02078 source
Section 4.1 compares recurrent networks with one, two, and three layers on two sequence datasets and states that a depth of at least two is consistently beneficial. This directly supports the attributed finding, within the paper’s tested settings. (LLM, conf=0.94)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“This seems to indicate that although deep learning techniques are often employed with large amounts of data, (e.g., millions of frames in computer vision [22]), they may actually be applicable to problem domains where acquiring annotated data is very costly, such as in supervised activity recognition.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 5. Results and Discussion > 5.4. Discussion
What the sources say
- Supportedcites [22] Pigou L. Oord A.V.D. Dieleman S. van Herreweghe M. Dambre J. Beyond Temporal Pooling: Recurrence and Temporal Convolutions for Gesture Recognition in Video arXiv Preprint 2015 arXiv:1506.01911 source
The citation is attached to the example of large-scale computer-vision data, not to the citing authors’ conclusion about costly annotation. Section 4.1 (page unavailable) reports 1,720,800 labeled video frames in the gesture-recognition dataset, supporting the cited example of data on the order of millions of frames. (LLM, conf=0.86)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The baseline CNN also offers better results than the OPPORTUNITY submissions in the recognition of models of locomotion.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 5. Results and Discussion > 5.1. Performance Comparison
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“From the results, we can also see how the model is able to learn from signals obtained from accelerometers, gyroscopes and magnetometers, fusing them without requiring any specific preprocessing.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 6. Conclusions
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“(iv) it is applicable even if the gestures are longer than the observation window.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 5. Results and Discussion > 5.4. Discussion
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
84 more claims
Create a free account to see all 122 claims and to verify the 2 references still open.
- It may state two separate claims as one. 14
- Probably not a substantial claim. 23
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?
The cited source doesn’t support the claim as written.
In the citing paper
“However, these complex network framework suffered from low ef ciency and can hardly meet real-time requirements in practice applications.”
From InnoHAR: A Deep Neural Network for Complex Human Activity Recognition (2019) · cites this paper as [19]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe cited source doesn’t support the claim as written.
In the citing paper
“However, due to the lack of filtration techniques applied to multimodal data, the system’s performance was low.”
From A Multimodal IoT-Based Locomotion Classification System Using Features Engineering and Recursive Neural Network (2023) · cites this paper as [15]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“and yield better performances than pure CNNs or LSTM networks [12]”
From Comparison of Feature Learning Methods for Human Activity Recognition Using Wearable Sensors (2018) · cites this paper as [12]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The proposed method outperforms other methods, such as those based on deep believe networks (DBNs) [10], SVM [10], and CNNs [13].”
From Deep Recurrent Neural Networks for Human Activity Recognition (2017) · cites this paper as [13]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The popular DeepConvLSTM model [7] can only afford four, moderately-sized convolution layers to learn effective data representations, and two LSTM layers for sequence modeling.”
From Complex Deep Neural Networks from Large Scale Virtual IMU Data for Effective Human Activity Recognition Using Wearables (2021) · cites this paper as [7]
Experimental — accuracy not yet measured
20 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture
“The length of the kernels is the same for every convolutional layer, being defined as Pl=5,∀l=2,⋯,5.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 4. Experimental Setup > 4.2. Performance Measure
“Following the OPPORTUNITY challenge experimental setup, the length of the window is 500 ms, with a step size of 250 ms.”
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“| Total | 996,800 +(128×nc)+nc | 996,800 +(128×nc)+nc | 7,443,136+(128×nc)+nc | 7,443,136+(128×nc)+nc |”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.3. Model Implementation and Training
Reported 996800 · recomputed 1037888
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“| 6 | Wai,Waf,Wac,Wao: 7232×128 | 942,592 | W: 57,856×128 | 7,405,696 |”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.3. Model Implementation and Training
Reported 942592 · recomputed 3702784
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“| 7 | Wai,Waf,Wac,Wao: 128×128 | 33,280 | W: 128×128 | 16,512 |”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.3. Model Implementation and Training
Reported 33280 · recomputed 131072
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562 · cited 1×
“Other publications showed much deeper networks, such as “GoogLeNet”, which is a 27-layer deep neural network applied to image classification [45].”
BackgroundReducing the dimensionality of data with neural networks
Hinton, G. E., & Salakhutdinov, R. R. (2006) · Science · doi:10.1126/science.1127647 · Cited by 20,928
Speech recognition with deep recurrent neural networks
Graves, A., Mohamed, A.-R., & Hinton, G. (2013) · 2013 IEEE International Conference on Acoustics, Speech and Signal Processing · doi:10.1109/icassp.2013.6638947 · Cited by 8,893 · cited 1×
“Networks using LSTM cells have offered better performance than standard recurrent units in speech recognition, where they gave state-of-the-art results in phoneme recognition [20].”
BackgroundConvolutional networks for images, speech, and time series
LeCun, Y., & Bengio, Y. (1998) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.5555/303568.303704 · Cited by 4,231
The class imbalance problem: A systematic study1
Japkowicz, N., & Stephen, S. (2002) · Intelligent Data Analysis · doi:10.3233/ida-2002-6504 · Cited by 3,295
DeepPose: Human Pose Estimation via Deep Neural Networks
Toshev, A., & Szegedy, C. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.214 · Cited by 3,185 · cited 1×
“Deep CNNs have had a major impact on fields, like content recommendation [24], speech recognition [25] and in computer vision [26–28], where they have become the de facto standard approach.”
BackgroundOverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
Sermanet, P., Eigen, D., Zhang, X., Mathieu, M., Fergus, R., & LeCun, Y. (2013) · arXiv · doi:10.48550/arxiv.1312.6229 · Cited by 2,680 · cited 1×
“Deep CNNs have had a major impact on fields, like content recommendation [24], speech recognition [25] and in computer vision [26–28], where they have become the de facto standard approach.”
BackgroundLee, H., Grosse, R., Ranganath, R., & Ng, A. Y. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553453 · Cited by 2,412 · cited 1×
“Deep learning techniques have outperformed many conventional methods in computer vision [13] and audio classification [14].”
MethodA review of wearable sensors and systems with application in rehabilitation
Patel, S., Park, H., Bonato, P., Chan, L., & Rodgers, M. (2012) · Journal of NeuroEngineering and Rehabilitation · doi:10.1186/1743-0003-9-21 · Cited by 2,396 · cited 1×
“, from simple hand gestures to complex activities, such as “cooking a meal”) and the context in which they occur from sensor data is at the core of smart assistive technologies, such as in smart homes [1], in rehabilitation [2], in health support [3,4], in skill assessment [5] or in industrial settings [6].”
Background
Research works printed but not linked: 1
- Gers F.A. Schraudolph N.N. Schmidhuber J. Learning precise timing with LSTM recurrent networks J. Mach. Learn. Res. 2003 3 115 143
Cited by
Items 701–800 of 2,807
Learning Disentangled Behaviour Patterns for Wearable-based Human Activity Recognition
Su, J., Wen, Z., Lin, T., & Guan, Y. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3517252 · Cited by 43
“CNN[56] and DeepConvLSTM[37] were the state-of-the-art feature learning approaches for human activity recognition; beta-VAE[17] was a conventional disentangle learning framework in the computer vision field and GILE [46] was a recent method aiming at disentangling the identity information from raw data streams, yet su…”
BackgroundMethodInfluentialA topical review on AI-interlinked biodomain sensors for multi-purpose applications
Thapa, R., Poudel, S., Krukiewicz, K., & Kunwar, A. (2024) · Measurement · doi:10.1016/j.measurement.2024.114123 · Cited by 38
Stereotypical Motor Movement Detection in Dynamic Feature Space
Rad, N. M., Kia, S. M., Zarbo, C., Jurman, G., Venuti, P., & Furlanello, C. (2016) · 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW) · doi:10.1109/icdmw.2016.0076 · Cited by 29
“Ordonez and Roggen [38] applied the deep architecture, by combining CNN with LSTM network, on human activity recognition problems.”
BackgroundZilelioglu, H., Khodabandelou, G., Chibani, A., & Amirat, Y. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3267243 · Cited by 32
“Luwe et al. [26] proposed replacing the LSTM of the DeepConvLSTM [5] with a bidirectional LSTM.”
BackgroundMethodInfluentialOffloading and Transmission Strategies for IoT Edge Devices and Networks
Kang, J., & Eom, D.-S. (2019) · Sensors · doi:10.3390/s19040835 · Cited by 32
Sensor Data Required for Automatic Recognition of Athletic Tasks Using Deep Neural Networks
Clouthier, A. L., Ross, G. B., & Graham, R. B. (2020) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2019.00473 · Cited by 29
“The network combines convolutional and recurrent layers (Ordóñez and Roggen, 2016).”
BackgroundMethodInfluentialBarua, A., Zhang, Z., Al-Turjman, F., & Yang, X. (2020) · IEEE Access · doi:10.1109/access.2020.3000599 · Cited by 30
“In [71], they used CNN with position-specific scoring matrices (PSSM) to recognize electron transport proteins and achieved an accuracy of 92.3%.”
MethodInfluentialHDL: Hierarchical Deep Learning Model based Human Activity Recognition using Smartphone Sensors
Su, T., Sun, H., Ma, C., Jiang, L., & Xu, T. (2019) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2019.8851889 · Cited by 33
“In [28], the authors proposed a generic deep framework (DeepConvLSTM) for activity recognition based on convolutional and LSTM recurrent units, which was suitable for multimodal wearable sensors.”
BackgroundAdaimi, R., & Thomaz, E. (2022) · Sensors · doi:10.3390/s22186881 · Cited by 35
“DeepConvLSTM [36], which has been widely used for HAR datasets, uses a combination of convolutional and recurrent layers.”
MethodYoshimura, N., Maekawa, T., Hara, T., Wada, A., & Namioka, Y. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3534572 · Cited by 38
“In particular, DeepConvLSTM proposed by Ordóñez et al. [35] has been used as a baseline in many activity recognition studies.”
BackgroundMethodInfluentialŞimşek, A. İ., Bulut, E., Gür, Y. E., & Tarla, E. G. (2024) · Energy · doi:10.1016/j.energy.2024.133102 · Cited by 31
Ashfaque Mostafa, T., Soltaninejad, S., McIsaac, T. L., & Cheng, I. (2021) · Sensors · doi:10.3390/s21196446 · Cited by 32
Learning multi-level features for sensor-based human action recognition
Xu, Y. L., Shen, Z., Zhang, X., Gao, Y., Deng, S., Wang, Y., Fan, Y., & Chang, E. (2017) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2017.07.001 · Cited by 32
“Instead of dealing with sparsely labeled cases, MLPL proposed in this paper implements MIL to learn discriminative latent patterns of actions, by which high-level features would be acquired.”
MethodWang, H., & Wang, L. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.02581 · Cited by 29
Spatiotemporal neural networks for action recognition based on joint loss
Jing, C., Wei, P., Sun, H., & Zheng, N. (2019) · Neural Computing and Applications · doi:10.1007/s00521-019-04615-w · Cited by 29
A benchmark for domain adaptation and generalization in smartphone-based human activity recognition
Napoli, O. O., Duarte, D., Alves, P., Soto, D. H. P., de Oliveira, H. E., Rocha, A. D. R., Boccato, L., & Borin, E. (2024) · Scientific Data · doi:10.1038/s41597-024-03951-4 · Cited by 52
Shetty, S., Ananthanarayana, V. S., & Mahale, A. R. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-14940-x · Cited by 36
OpenPack: A Large-Scale Dataset for Recognizing Packaging Works in IoT-Enabled Logistic Environments
Yoshimura, N., Morales, J., Maekawa, T., & Hara, T. (2024) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom59722.2024.10494448 · Cited by 31
“For example, when two types of usages of IoT-enabled devices are available, e.g., scanning with a handheld scanner and printing with a label printer, N A is 2.”
BackgroundHuman Action Recognition Using CNN-SVM Model
Athavale, V. A., Gupta, S. C., Kumar, D., & Savita, S. (2021) · Advances in science and technology · doi:10.4028/www.scientific.net/ast.105.282 · Cited by 36
“IEEE, [11] Ordóñez, Francisco Javier, and Daniel Roggen 2016.”
BackgroundA Comprehensive Methodological Survey of Human Activity Recognition Across Diverse Data Modalities
Shin, J., Hassan, N., Miah, A. S. M., & Nishimura, S. (2025) · Sensors · doi:10.3390/s25134028 · Cited by 58
Weakly Supervised Multi-Task Representation Learning for Human Activity Analysis Using Wearables
Sheng, T., & Huber, M. (2020) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3397330 · Cited by 28
Liu, T., Wang, S., Liu, Y., Quan, W., & Zhang, L. (2021) · The Journal of Supercomputing · doi:10.1007/s11227-021-04140-5 · Cited by 32
Online Evolving Interval Type-2 Intuitionistic Fuzzy LSTM-Neural Networks for Regression Problems
Wang, Y., & Luo, C. (2019) · IEEE Access · doi:10.1109/access.2019.2904630 · Cited by 30
“[46] proposed a general depth framework to implement public activity identification.”
BackgroundFalco, I. D., Pietro, G. D., & Sannino, G. (2019) · Neural Computing and Applications · doi:10.1007/s00521-018-03973-1 · Cited by 31
RetractedFinger-Gesture Recognition for Visible Light Communication Systems Using Machine Learning
Webber, J. L., Mehbodniya, A., Teng, R., Arafa, A., & Alwakeel, A. S. (2021) · Applied Sciences · doi:10.3390/app112411582 · Cited by 28
“The long short-term memory (LSTM) algorithm is a type of recurrent neural network that can efficiently learn time-series sequences that are increasingly used in ML-based HAR systems, such as [8], for wearable activity recognition [9] and sign language translation [10].”
BackgroundYin, C., Miao, X., Chen, J., Jiang, H., Chen, D., Tong, Y., & Zheng, S. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3243944 · Cited by 31
“Researchers utilized smartphones or wearable devices to recognize human activities in [8], [9], [10], [11], [12], and [13], but they were not always carrying equipment.”
BackgroundAn Energy-Efficient Fall Detection Method Based on FD-DNN for Elderly People
Liu, L., Hou, Y., He, J., Lungu, J., & Dong, R. (2020) · Sensors · doi:10.3390/s20154192 · Cited by 30
Deep learning with 4D spatio-temporal data representations for OCT-based force estimation
Gessert, N., Bengs, M., Schlüter, M., & Schlaefer, A. (2020) · Medical Image Analysis · doi:10.1016/j.media.2020.101730 · Cited by 31
Mangal, N. K., & Tiwari, A. K. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104316 · Cited by 28
Improving CNN-based activity recognition by data augmentation and transfer learning
Kalouris, G., Zacharaki, E. I., & Megalooikonomou, V. (2019) · 2019 IEEE 17th International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin41052.2019.8972135 · Cited by 28
“1% and 90% depending on the context and method applied, while the accuracy is increased in a more recent work implementing deep convolutional and LSTM recurrent neural networks [17].”
BackgroundMethodHwang, I., Cha, G., & Oh, S. (2017) · 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI) · doi:10.1109/mfi.2017.8170441 · Cited by 28
“Furthermore, a neural network is exploited for better classification without putting the effort of humans for designing proper features [4].”
BackgroundStudy of assessment of cognitive ability of human brain using deep learning
Sinha, G. R. (2017) · International Journal of Information Technology · doi:10.1007/s41870-017-0025-8 · Cited by 32
“Ordonez and Roggen [10] studied deep non-recurrent networks for human activity recognition and the work outperformed some of the previous works by 9%.”
ResultProgress of Human Action Recognition Research in the Last Ten Years: A Comprehensive Survey
Singh, P. K., Kundu, S., Adhikary, T., Sarkar, R., & Bhattacharjee, D. (2021) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-021-09681-9 · Cited by 37
Do we need early exit networks in human activity recognition?
Lattanzi, E., Contoli, C., & Freschi, V. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106035 · Cited by 33
Varshney, N., Bakariya, B., & Kushwaha, A. K. S. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11131-4 · Cited by 30
Qiao, Y., Clark, C., Lomax, S., Kong, H., Su, D., & Sukkarieh, S. (2021) · Frontiers in Animal Science · doi:10.3389/fanim.2021.759147 · Cited by 42
“Time-series data such as video, on the other hand, contain more temporal information and have been widely used to monitor cattle behaviors and welfare (Ordóñez and Roggen, 2016; Bahlo et al., 2019).”
BackgroundAnalysis of Multi-Sensor Fusion for Mobile and Wearable Sensor Based Human Activity Recognition
Nweke, H. F., Teh, Y. W., Alo, U. R., & Mujtaba, G. (2018) · Proceedings of the International Conference on Data Processing and Applications · doi:10.1145/3224207.3224212 · Cited by 28
“in the data and replace missing values with the previous sensor values in the data [13].”
MethodLarge Language Models are Zero-Shot Recognizers for Activities of Daily Living
Civitarese, G., Fiori, M., Choudhary, P., & Bettini, C. (2024) · arXiv · doi:10.48550/arxiv.2407.01238 · Cited by 28
A hybridization of deep learning techniques to predict and control traffic disturbances
Louati, A. (2020) · Artificial Intelligence Review · doi:10.1007/s10462-020-09831-8 · Cited by 28
“Ordóñez et al. (2016) have proved that LSTM has superior performance in anomaly detection.”
BackgroundCross-dataset deep transfer learning for activity recognition
Gjoreski, M., Kalabakov, S., Luštrek, M., Gams, M., & Gjoreski, H. (2019) · Adjunct Proceedings of the 2019 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2019 ACM International Symposium on Wearable Computers · doi:10.1145/3341162.3344865 · Cited by 28
Displacement prediction of water-induced landslides using a recurrent deep learning model
Meng, Q., Wang, H., He, M., Gu, J., Qi, J. F., & Yang, L. (2020) · European Journal of Environmental and Civil engineering · doi:10.1080/19648189.2020.1763847 · Cited by 28
Identifying and Monitoring the Daily Routine of Seniors Living at Home
Chifu, V. R., Pop, C. B., Demjen, D., Socaci, R., Todea, D., Antal, M., Cioara, T., Anghel, I. M., & Antal, C. (2022) · Sensors · doi:10.3390/s22030992 · Cited by 35
Design of Motor Skill Recognition and Hierarchical Evaluation System for Table Tennis Players
Shi, Z., Jia, Y., Shi, G., Zhang, K., Ji, L., Wang, D., & Wu, Y. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3346880 · Cited by 38
“Human activity recognition for wearable devices is an emerging research field, which focuses on the use of pattern recognition technology in wearable devices [20], [21].”
BackgroundAnkalaki, S., & Thippeswamy, M. N. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15766-3 · Cited by 31
Wang, C., Duan, Q., Zhou, K., et al. (2020) · Acta Physica Sinica · doi:10.7498/aps.69.20191935 · Cited by 27
Kim, K., & Cho, Y. K. (2020) · Journal of Construction Engineering and Management · doi:10.1061/(asce)co.1943-7862.0002001 · Cited by 32
Towards deep clustering of human activities from wearables
Abedin, A., Motlagh, F., Shi, Q., Rezatofighi, H., & Ranasinghe, D. (2020) · Proceedings of the 2020 International Symposium on Wearable Computers · doi:10.1145/3410531.3414312 · Cited by 27
“Researchers have explored CNNs [8, 33, 39–41], RNNs [15, 18], and a combination of convolutional and recurrent layers [1, 29, 31] to effectively model the temporal dependencies inherent in sequences captured with sensors.”
BackgroundAirScript - Creating Documents in Air
Dash, A., Sahu, A., Shringi, R., Gamboa, J. C. B., Afzal, M. Z., Malik, M. I., Dengel, A., & Ahmed, S. (2017) · IAPR International Conference on Document Analysis and Recognition (ICDAR) · doi:10.1109/icdar.2017.153 · Cited by 36
“Similar Deep Learning models, such as Convolutional Neural Networks [18] and LSTMs [19], have also shown improved results in the domain of gesture recognition.”
ResultReal-time multimodal interaction in virtual reality - a case study with a large virtual interface
Cao, L., Zhang, H., Peng, C. Y., & Hansberger, J. T. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-14381-6 · Cited by 41
“Deep learning models were proposed for fusing temporal multimodal information with the praised capability to learn from a large number of data [10, 12, 17, 27, 28].”
BackgroundIntelligent Brushing Monitoring Using a Smart Toothbrush with Recurrent Probabilistic Neural Network
Chen, C.-H., Wang, C.-C., & Chen, Y.-Z. (2021) · Sensors · doi:10.3390/s21041238 · Cited by 39
“The Convolutional Neural Network (CNN) and the Long Short-Term Memory (LSTM) model are used to identify sensor data [39,40].”
MethodHan, T., & Sánchez-Azofeifa, A. G. (2022) · Remote Sensing · doi:10.3390/rs14133157 · Cited by 32
Toward Automated Feature Extraction for Deep Learning Classification of Electrocardiogram Signals
Butt, F. S., Wagner, M., Schäfer, J., & Gómez‐Ullate, D. (2022) · IEEE Access · doi:10.1109/access.2022.3220670 · Cited by 31
“Ordóñez et al. [15] proposed an activity recognition classifier that combines a deep CNN and dense layers.”
BackgroundFeng, H., Shen, Q., Song, R., Shi, L., & Xu, H. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121296 · Cited by 32
Machine Learning for Detection and Risk Assessment of Lifting Action
Thomas, B., Lu, M., Jha, R., & Bertrand, J. (2022) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2022.3212666 · Cited by 30
“Table V compares the performance of the proposed model to other baseline models [6], [7], [12], as well as a standard linear kernel SVM.”
MethodModeling transformer architecture with attention layer for human activity recognition
Pareek, G., Nigam, S., & Singh, R. K. (2024) · Neural Computing and Applications · doi:10.1007/s00521-023-09362-7 · Cited by 28
Human Activity Recognition Based on Hybrid Learning Algorithm for Wearable Sensor Data
Athota, R. K., & Sumathi, D. (2022) · Measurement Sensors · doi:10.1016/j.measen.2022.100512 · Cited by 48
Estimation of Hand Motion from Piezoelectric Soft Sensor Using Deep Recurrent Network
Kim, S. H., Kwon, Y., Kim, K., & Cha, Y. (2020) · Applied Sciences · doi:10.3390/app10062194 · Cited by 27
An Artificial Intelligence Perspective on Mobile Edge Computing
Chen, Z., He, Q., Liu, L., Lan, D., Chung, H.-M., & Mao, Z. (2019) · IEEE International Conference on Smart Internet of Things (SmartIoT) · doi:10.1109/smartiot.2019.00024 · Cited by 26
“was introduced in order to control the access to memory cells, which has been used in many applications successfully [32] [33] [34] [35].”
MethodBehavior Classification and Spatiotemporal Analysis of Grazing Sheep Using Deep Learning
Jin, Z., Shu, H., Hu, T., Jiang, C., Yan, R., Qi, J., Wang, W., & Guo, L. (2023) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.108894 · Cited by 26
Human Activity Recognition based on Real Life Scenarios
Machot, F. A., Ranasinghe, S., Plattner, J., & Jnoub, N. (2018) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomw.2018.8480138 · Cited by 27
“Also, [9] suggests a method to automate the feature extraction process from raw sensor data using a deep convolutional neural network.”
MethodDeep Learning for Automated Occlusion Edge Detection in RGB-D Frames
Sarkar, S., Venugopalan, V., Reddy, K., Ryde, J., Jaitly, N., & Giering, M. J. (2016) · Journal of Signal Processing Systems · doi:10.1007/s11265-016-1209-3 · Cited by 26
AI Benchmark: Running Deep Neural Networks on Android SmartphonesPreprint
Ignatov, A., Timofte, R., Chou, W., Wang, K., Wu, M., Hartley, T., & Gool, L. V. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.01109 · Cited by 82
“A separte group deals with on-line sensor data processing for human activity recognition from accelerometer data [25, 26], gesture recognition [27] or sleep monitoring [28].”
BackgroundAn Improved Deep Convolutional LSTM for Human Activity Recognition Using Wearable Sensors
Zhang, N., Song, Y., Fang, D., Gao, Z., & Yan, Y. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3335213 · Cited by 36
“The recurrent layers mainly include two recurrent LSTM layers with 128 cells, and the number of recurrent layers is determined following the results presented in [36].”
A Novel Signal Localized Convolution Neural Network for Power Transformer Differential Protection
Murugan, S. K., Simon, S. P., & Eapen, R. R. (2021) · IEEE Transactions on Power Delivery · doi:10.1109/tpwrd.2021.3080927 · Cited by 25
HierHAR: Sensor-Based Data-Driven Hierarchical Human Activity Recognition
Wang, A., Zhao, S., Zheng, C., Chen, H., Liu, L., & Chen, G. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3023860 · Cited by 29
“For the performance evaluation metrics, we use accuracy ( Acc ), precision ( Prec ), recall ( Rec ), F 1, and G-mean ( Gm ) to evaluate the activity recognizers [43].”
BackgroundMethodHuman Activity Recognition by Using Different Deep Learning Approaches for Wearable Sensors
Erdaş, Ç. B., & Güney, S. (2021) · Neural Processing Letters · doi:10.1007/s11063-021-10448-3 · Cited by 33
Bian, S., Liu, M., Zhou, B., Lukowicz, P., & Magno, M. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3643555 · Cited by 32
“Since human activities are made of complex sequences of motor movements, and capturing these temporal dynamics is fundamental for successful HAR [120], Recurrent Neural Network is popular for such recognition tasks.”
BackgroundAre Microcontrollers Ready for Deep Learning-Based Human Activity Recognition?
Elsts, A., & McConville, R. (2021) · Electronics · doi:10.3390/electronics10212640 · Cited by 33
Pre-ignition Detection Using Deep Neural Networks: A Step Towards Data-driven Automotive Diagnostics
Wolf, P., Mrowca, A., Nguyen, T. T., Bäker, B., & Günnemann, S. (2018) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2018.8569908 · Cited by 36
“Referring to [32] and [26], the shorthand description of DADN is: C(64)− C(64)− C(64)− C(64)− R(96)− R(96)− Sm, where C(F ) is a convolutional layer l with F l feature maps, R(z) is a recurrent LSTM layer l with z cells, and Sm is a softmax layer.”
BackgroundInfluentialsEMG-Based Upper Limb Elbow Force Estimation Using CNN, CNN-LSTM, and CNN-GRU Models
Wahid, A., Ullah, K., Ullah, S. I., et al. (2024) · IEEE Access · doi:10.1109/access.2024.3451209 · Cited by 26
Health Monitoring of Automotive Suspensions: A LSTM Network Approach
Hu, H., Luo, H., & Deng, X. (2021) · Shock and Vibration · doi:10.1155/2021/6626024 · Cited by 35
Nweke, H. F., Al-Garadi, M. A., Mujtaba, G., Alo, U. R., & Waqas, A. (2018) · International Conference on Computing, Mathematics and Engineering Technologies (iCoMET) · doi:10.1109/icomet.2018.8346364 · Cited by 31
“The architecture is similar to the one proposed by [16] but instead of Long Short Term Memory (LSTM), we introduce the Gated Recurrent unit as shown in Fig 2 below.”
BackgroundMethodInfluentialAI-Based Smart Sensing and AR for Gait Rehabilitation Assessment
Monge, J., Ribeiro, G., Raimundo, A., Postolache, O. A., & dos Santos, J. A. F. (2023) · Information · doi:10.3390/info14070355 · Cited by 43
“These systems typically use sensors, cameras, or wearable devices to collect data and employ machine learning algorithms to recognize and interpret the patterns of daily living [44–46].”
BackgroundInvestigating Enhancements to Contrastive Predictive Coding for Human Activity Recognition
Haresamudram, H., Essa, I., & Plötz, T. (2023) · 2023 IEEE International Conference on Pervasive Computing and Communications (PerCom) · doi:10.1109/percom56429.2023.10099197 · Cited by 27
“The supervised baselines comprise DeepConvLSTM [51], an established network for recognizing activities, a Conv.”
BackgroundPatchHAR: A MLP-Like Architecture for Efficient Activity Recognition Using Wearables
Wang, S., Zhang, L., Wang, X., Huang, W., Wu, H., & Song, A. (2024) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2024.3354261 · Cited by 33
“For fair comparisons, we strictly follow the same split strategy available in recent research literature [19], [20], [36], [37].”
MethodComparison of machine learning approaches for near-fall-detection with motion sensors
Hellmers, S., Krey, E., Gashi, A., Koschate, J., Schmidt, L., Stuckenschneider, T., Hein, A., & Zieschang, T. (2023) · Frontiers in Digital Health · doi:10.3389/fdgth.2023.1223845 · Cited by 27
“Also a deep convolutional and LSTM recurrent neural network (DeepConvLSTM) was used, which combines the advantages of CNNs for feature extraction and LSTM neurons for pattern recognition from the temporal sequence of the data (12).”
MethodErcolano, G., Riccio, D., & Rossi, S. (2017) · IEEE RO-MAN · doi:10.1109/roman.2017.8172406 · Cited by 27
“that concern spatio-temporal classification like in [13] for video description and in [14] for activity recognition from wearable devices data.”
BackgroundSALON: Simplified Sensing System for Activity of Daily Living in Ordinary Home
Matsui, T., Onishi, K., Misaki, S., Fujimoto, M., Suwa, H., & Yasumoto, K. (2020) · Sensors · doi:10.3390/s20174895 · Cited by 27
Gesture Prediction Using Wearable Sensing Systems with Neural Networks for Temporal Data Analysis
Kanokoda, T., Kushitani, Y., Shimada, M., & Shirakashi, J. (2019) · Sensors · doi:10.3390/s19030710 · Cited by 26
Perez-Gamboa, S., Sun, Q., & Yan, Z. (2021) · IEEE International Symposium on Inertial Sensors and Systems (INERTIAL) · doi:10.1109/inertial51137.2021.9430460 · Cited by 34
“A deep convolutional and recurrent model referred to as “DeepConvLSTM” was presented in [17], which achieved a higher HAR rate than a baseline CNN model.”
MethodChhowa, T. T., Rahman, M. A., Paul, A. K., & Ahmmed, R. (2019) · International Conference on Electrical, Computer and Communication Engineering · doi:10.1109/ecace.2019.8679200 · Cited by 30
“When there is no time limitation, LSTM models perform better than recurrent neural network models [16] due to an efficient prediction of the target based on proper data points.”
InfluentialOleh, U., Obermaisser, R., & Ahammed, A. S. (2024) · Algorithms · doi:10.3390/a17100434 · Cited by 40
Adversarial Deep Feature Extraction Network for User Independent Human Activity Recognition
Suh, S., Rey, V. F., & Lukowicz, P. (2022) · 2022 IEEE International Conference on Pervasive Computing and Communications (PerCom) · doi:10.1109/percom53586.2022.9762387 · Cited by 25
Improved convolutional neural network combined with rough set theory for data aggregation algorithm
Cao, J., Zhang, X., Zhang, C., et al. (2018) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-018-1068-9 · Cited by 25
“These filters convolute with input data and obtain an output, called as feature maps (Ordonez and Roggen 2016).”
BackgroundMethodWiFi-based indoor passive fall detection for medical Internet of Things
Xia, Z., & Chong, S. (2023) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2023.108763 · Cited by 26
Mohammed, S., & Tashev, I. (2017) · 2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks (BSN) · doi:10.1109/bsn.2017.7936037 · Cited by 29
“Among the sensors available, the authors in [9] utilized 113 channels: 3 channels of accelerometer data from 12 locations, 9 channels of IMU data from 5 locations and 16 channels from 2 sensors measuring Euler orientations and angular velocity.”
Yao, M., Zhang, L., Cheng, D., Qin, L., Liu, X., Fu, Z., Wu, H., & Song, A. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3371462 · Cited by 31
“Furthermore, Ordóñez and Roggen [21] fused CNNs and LSTM to integrate spatial and temporal features, improving the accuracy and robustness of activity recognition through multimodal data fusion.”
BackgroundInfluentialADAM-sense: Anxiety-displaying activities recognition by motion sensors
Khan, N. S., Ghani, M. S., & Anjum, G. (2021) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2021.101485 · Cited by 25
SwallowNet: Recurrent neural network detects and characterizes eating patterns
Nguyen, D. T., Cohen, E., Pourhomayoun, M., & Alshurafa, N. (2017) · 2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomw.2017.7917596 · Cited by 25
Renani, M. S., Mahoor, M. H., & Clary, C. W. (2023) · Sensors · doi:10.3390/s23135778 · Cited by 33
Hand Gesture Recognition in Video Sequences Using Deep Convolutional and Recurrent Neural Networks
Obaid, F., Babadi, A., & Yoosofan, A. (2020) · Applied Computer Systems · doi:10.2478/acss-2020-0007 · Cited by 25
“Deep learning has been decreasing reliance on engineered features to address increasingly complex recognition problems [9].”
BackgroundAutoAugHAR: Automated Data Augmentation for Sensor-based Human Activity Recognition
Zhou, Y., Zhao, H., Huang, Y., Röddiger, T., Kurnaz, M., Riedel, T., & Beigl, M. (2024) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3659589 · Cited by 25
“To improve the capacity to extract temporal information, the hybrid model DeepConvLSTM ( DCL ) [51] incorporates LSTM layers into the individual convolution network.”
BackgroundMethodFall Detection with Supervised Machine Learning using Wearable Sensors
Giuffrida, D., Benetti, G., Martini, D. D., & Facchinetti, T. (2019) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin41052.2019.8972246 · Cited by 29
“The authors of [11] utilize a Deep Learning (DL) technique to detect falls.”
MethodHuman Activity Recognition Using 1D Convolutional Neural Network
Banjarey, K., Sahu, S. P., & Dewangan, D. K. (2021) · Advances in intelligent systems and computing · doi:10.1007/978-981-16-5157-1_54 · Cited by 28
Run-Time Efficient RNN Compression for Inference on Edge Devices
Thakker, U., Beu, J., Gope, D., Dasika, G., & Mattina, M. (2019) · 2019 2nd Workshop on Energy Efficient Machine Learning and Cognitive Computing for Embedded Applications (EMC2) · doi:10.1109/emc249363.2019.00013 · Cited by 24
“The second HAR network is based on the work in [12].”
BackgroundA benchmark for computational analysis of animal behavior, using animal-borne tags
Hoffman, B., Cusimano, M., Baglione, V., Canestrari, D., Chevallier, D., DeSantis, D. L., Jeantet, L., Ladds, M., Maekawa, T., Mata‐Silva, V., et al. (2024) · Movement Ecology · doi:10.1186/s40462-024-00511-8 · Cited by 37
Estimating Visual Focus of Attention in Multiparty Meetings using Deep Convolutional Neural Networks
Otsuka, K., Kasuga, K., & Martina, K. (2018) · INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION · doi:10.1145/3242969.3242973 · Cited by 27
“It would also be worth trying other DNN models, such as DeepCon-vLSTM [27], that outperform CNNs in HAR tasks.”
MethodSelf-supervised Learning for Accelerometer-based Human Activity Recognition: A Survey
Logacjov, A. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3699767 · Cited by 40
A Hybrid Deep Learning System for Real-World Mobile User Authentication Using Motion Sensors
Zhu, T., Weng, Z., Chen, G., & Fu, L. (2020) · Sensors · doi:10.3390/s20143876 · Cited by 28
Bargellesi, N., Carletti, M., Cenedese, A., Susto, G. A., & Terzi, M. (2019) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2019.09.129 · Cited by 25
Metadata
- Authors
- Ordóñez, F., & Roggen, D.
- Affiliations
- University of Sussex (GB)
- Corresponding author
- Ordóñez, Francisco
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2016-01-18
- Volume
- 16
- Issue
- 1
- Pages
- 115
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC4732148
- OpenAlex
- W2270470215
- Cited by
- 2,761
- References
- 49
- Citations by year
- 2026: 284 · 2025: 256 · 2024: 284 · 2023: 270 · 2022: 319 · 2021: 379 · 2020: 347 · 2019: 288 · 2018: 219 · 2017: 93 · 2016: 21
- FWCI
- 140.52
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Sussex Research Online (University of Sussex) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by)
- Funders
- Google Faculty Research Award (Is deep learning useful for wearable activity recognition?)
- EPSRC (EP/N007816/1)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Engineering › Biomedical Engineering › Gait Recognition and Analysis
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Keywords
- human activity recognition, wearable sensors, multimodal sensor fusion, convolutional neural networks, LSTM neural network, temporal dynamics, automatic feature extraction, hyperparameter optimization
- MeSH terms
- Machine Learning, Clothing, Humans, Human Activities, Signal Processing, Computer-Assisted, Databases, Factual, Neural Networks, Computer, Monitoring, Ambulatory
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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.