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 401–500 of 2,807
Time series classification using local distance-based features in multi-modal fusion networks
Iwana, B. K., & Uchida, S. (2019) · Pattern Recognition · doi:10.1016/j.patcog.2019.107024 · Cited by 58
“Fusion networks have also been used for combining title and image data [25] and for multi-source data types [27, 28], activities of daily life [27].”
MethodA Novel Distribution-Embedded Neural Network for Sensor-Based Activity Recognition
Qian, H., Pan, S. J., Da, B., & Miao, C. (2019) · Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2019/779 · Cited by 56
“The state-of-the-art for sensor-based activity recognition are basically combinations of these three types of base models [Morales and Roggen, 2016].”
BackgroundMethodInfluentialChakraborty, J., & Nandy, A. (2020) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2020.102076 · Cited by 62
Fall prediction using behavioural modelling from sensor data in smart homes
Forbes, G., Massie, S., & Craw, S. (2019) · Artificial Intelligence Review · doi:10.1007/s10462-019-09687-7 · Cited by 68
“Recurrent Neural Nets (RNN) employing Long Short-Term Memory (LSTM), offering the ability to model sequential data, have also been applied for human activity recognotion with data from wearables (Ordóñez and Roggen 2016).”
MethodExploring transformers for behavioural biometrics: A case study in gait recognition
Delgado-Santos, P., Tolosana, R., Guest, R., Deravi, F., & Vera-Rodríguez, R. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109798 · Cited by 74
Deep Learning in Gait Parameter Prediction for OA and TKA Patients Wearing IMU Sensors
Renani, M. S., Myers, C. A., Zandie, R., Mahoor, M. H., Davidson, B. S., & Clary, C. W. (2020) · Sensors · doi:10.3390/s20195553 · Cited by 60
Sensor-based activity recognition in the context of ambient assisted living systems: A review
Patel, A. D., & Shah, J. H. (2019) · Journal of Ambient Intelligence and Smart Environments · doi:10.3233/ais-190529 · Cited by 60
“Ordóñez and Roggen [103] proposed Long-short-term memory recurrent (LSTMs) neural networks (RNN).”
Deep Learning for SARS COV-2 Genome Sequences
Whata, A., & Chimedza, C. (2021) · IEEE Access · doi:10.1109/access.2021.3073728 · Cited by 62
Human Activity Recognition Using Temporal Convolutional Network
Nair, N., Thomas, C., & Jayagopi, D. B. (2018) · International Workshop on Sensor-based Activity Recognition and Interaction (iWOAR) · doi:10.1145/3266157.3266221 · Cited by 75
“Newer approaches like [22] [34] uses deep learning meth-ods for HAR.”
MethodDeep Learning for Action and Gesture Recognition in Image Sequences: A Survey
Asadi, M., Clapés, A., Bellantonio, M., Escalante, H. J., Ponce-López, V., Baró, X., Guyon, I., Kasaei, S., & Escalera, S. (2017) · The Springer series on challenges in machine learning · doi:10.1007/978-3-319-57021-1_19 · Cited by 67
Jaén-Vargas, M., Leiva, K. M. R., Fernandes, F., Gonçalves, S. B., Silva, M. T., Lopes, D. S., & Olmedo, J. J. S. (2022) · PeerJ Computer Science · doi:10.7717/peerj-cs.1052 · Cited by 83
DeActive: Scaling Activity Recognition with Active Deep Learning
Hossain, H. M. S., Al Haiz Khan, M. A., & Roy, N. (2018) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3214269 · Cited by 56
“Francisco et al [21] demonstrated a model based on deep convolution and LSTM recurrent network which is suitable for multimodal wearable sensors.”
BackgroundZhang, J., Wu, Z., Li, F., Luo, J., Ren, T., Hu, S., Li, W., & Li, W. (2019) · IEEE Access · doi:10.1109/access.2019.2932434 · Cited by 57
“Related work has used the weighted F1-score as primary performance metric previously [57].”
MethodFluid Intake Monitoring Systems for the Elderly: A Review of the Literature
Cohen, R., Fernie, G. R., & Fekr, A. R. (2021) · Nutrients · doi:10.3390/nu13062092 · Cited by 64
PSDRNN: An Efficient and Effective HAR Scheme Based on Feature Extraction and Deep Learning
Xiao, L., Wang, Y., Zhang, B., & Ma, J. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.2968920 · Cited by 62
“Our proposed PSDRNN and tri-PSDRNN schemes are free from heavy feature selection with the introduction of PSD. xyz -CNNDRNN HAR schemes [16], [25] first used the convolution-activation-pooling block in the CNN model to explicitly extract the virtual features, and then input these features to the…”
MethodDeep learning approach on information diffusion in heterogeneous networks
Molaei, S., Zare, H., & Veisi, H. (2019) · Knowledge-Based Systems · doi:10.1016/j.knosys.2019.105153 · Cited by 59
“In this section this method is briefly described just for a more detailed comparison with other introduced methods [56].”
MethodAsymmetric Residual Neural Network for Accurate Human Activity Recognition
Long, J., Sun, W., Yang, Z., & Raymond, O. I. (2019) · Information · doi:10.3390/info10060203 · Cited by 54
Convolutional long short term memory deep neural networks for image sequence prediction
Balderas, D., Ponce, P., & Molina, A. (2018) · Expert Systems with Applications · doi:10.1016/j.eswa.2018.12.055 · Cited by 56
Cai, G., Zheng, X., Guo, J. Z., & Gao, W. (2024) · Safety Science · doi:10.1016/j.ssci.2024.106690 · Cited by 63
Paschalides, D., Stefanidis, D., Andreou, A. S., Orphanou, K., Pallis, G., Dikaiakos, M. D., & Markatos, E. P. (2020) · ACM Transactions on Internet Technology · doi:10.1145/3371276 · Cited by 58
“In the literature, the CNN is considered to be effective in extracting abstract features from the input [26], whereas the RNN works well with orderly sequence learning problems [44].”
MethodUnsupervised Human Activity Recognition Using the Clustering Approach: A Review
Ariza-Colpas, P., Vicario, E., De-La-Hoz-Franco, E., Piñeres-Melo, M. A., Carrascal, A. I. O., & Patara, F. (2020) · Sensors · doi:10.3390/s20092702 · Cited by 66
“…line of work, a series of datasets have been built, which will be explained in more detail in the following sections, of which the following stand out: Kasteren [28], CASAS (Kyoto [29], Aruba [30], Multiresident [31]), UCI Human Activity Recognition (HAR) [32], Opportunity [33], and mHealth [34].”
BackgroundInfluentialCheng, D., Zhang, L., Bu, C., Wu, H., & Song, A. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110789 · Cited by 55
Novel approaches to human activity recognition based on accelerometer data
Jordão, A., Tôrres, L. A. B., & Schwartz, W. R. (2018) · Signal Image and Video Processing · doi:10.1007/s11760-018-1293-x · Cited by 61
“However, recent works have presented better results employing Convolutional Neural Network (ConvNets) as an alternative to handcrafted approaches [13, 24,31].”
ResultSpatiotemporal Modeling for Video Summarization Using Convolutional Recurrent Neural Network
Yuan, Y., Li, H., & Wang, Q. (2019) · IEEE Access · doi:10.1109/access.2019.2916989 · Cited by 58
“Reference [34] presents a generic deep framework for multimodal wearable activity recognition, which is base on CNN and LSTM, and the experiments demonstrate the satisfying efficacy of the framework.”
MethodUsing a deep learning network to recognise low back pain in static standing
Hu, B., Kim, C. H., Ning, X., & Xu, X. (2018) · Ergonomics · doi:10.1080/00140139.2018.1481230 · Cited by 62
“Previous studies have shown that deep learning networks with embedded long short-term memory (LSTM) units may achieve state-of-the art performance in motion category classification among healthy people (Ordonez and Roggen 2016).”
BackgroundSadhu, V., Anjum, K., & Pompili, D. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2023.3269380 · Cited by 60
“We do not use pooling layers after convolutional layers as our input data have been segmented into windows, and the output of the convolutional layers needs to be passed through time-series LSTM layers [54].”
MethodStoeve, M., Schuldhaus, D., Gamp, A., Zwick, C., & Eskofier, B. M. (2021) · Sensors · doi:10.3390/s21093071 · Cited by 63
“The received abstract representation of the input raw signal is consecutively modelled by recurrent layers [55].”
BackgroundCheng, W.-Y., Scotland, A., Lipsmeier, F., Kilchenmann, T., Jin, L., Schjodt-Eriksen, J., Wolf, D., Zhang-Schaerer, Y.-P., Garcia, I. F., Siebourg-Polster, J., et al. (2017) · 2017 IEEE/ACM International Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE) · doi:10.1109/chase.2017.87 · Cited by 54
“Similar structures have been used previously for HAR and have been shown to out-perform the traditional machine learning methods [3].”
Mekruksavanich, S., & Jitpattanakul, A. (2022) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2022265 · Cited by 59
Nouredanesh, M., Godfrey, A., Howcroft, J., Lemaire, E. D., & Tung, J. Y. (2020) · Gait & Posture · doi:10.1016/j.gaitpost.2020.04.010 · Cited by 63
Yang, P., Yang, C., Lanfranchi, V., & Ciravegna, F. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3142315 · Cited by 57
Wang, G., Wei, W., Jiang, J., Ning, C., Chen, H., Huang, J., Liang, B., Zang, N., Liao, Y., Chen, R., et al. (2019) · Epidemiology and Infection · doi:10.1017/s095026881900075x · Cited by 69
“However, the LSTM model can model the data of a time series, particularly the long-term data, and automatically determine the optimal time lag [33].”
MethodMcfly: Automated deep learning on time series
van Kuppevelt, D. E., Meijer, C., Huber, F., van der Ploeg, A., Georgievska, S., & van Hees, V. T. (2020) · SoftwareX · doi:10.1016/j.softx.2020.100548 · Cited by 55
Jin, X., Su, T., Kong, J., Bai, Y., Miao, B., & Dou, C. (2018) · Applied Sciences · doi:10.3390/app8030379 · Cited by 55
“Together with CNN, it can model the series-relationship of signals and achieves a good application effect in the recognition of the video captioning model [135].”
MethodOnline Recognition of Incomplete Gesture Data to Interface Collaborative Robots
Simao, M. A., Gibaru, O., & Neto, P. (2019) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2019.2891449 · Cited by 54
“Deep learning combined with recurrent LSTM networks demonstrated the state-of-the-art performance in the classification of human activities from wearable sensors [15].”
BackgroundAn overview of human activity recognition based on smartphone
Yuan, G., Wang, Z., Meng, F., Yan, Q., & Xia, S. (2018) · Sensor Review · doi:10.1108/sr-11-2017-0245 · Cited by 60
Issa, M., Helmi, A. M., Al‐qaness, M. A. A., Dahou, A., Elaziz, M. A., & Damaševičius, R. (2022) · Healthcare · doi:10.3390/healthcare10061084 · Cited by 64
Moreau, A., Anderer, P., Ross, M., Cerny, A., Almazan, T. H., & Peterson, B. (2018) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2017.2710798 · Cited by 53
“RNNs have been applied previously to accelerometer data in the context of gesture recognition in [22] (combining data from both an accelerometer and a gyrometer) or human activity recognition in [23], [24] (combining convolutional and recurrent units).”
MethodHandling annotation uncertainty in human activity recognition
Kwon, H., Abowd, G. D., & Plötz, T. (2019) · Proceedings of the 23rd International Symposium on Wearable Computers · doi:10.1145/3341163.3347744 · Cited by 52
“For this evaluation, we used the Opportunity dataset [6], and a state-of-the-art Con-vLSTM classifier [24], which not only automatically learns the feature representation but also models long-term temporal contexts.”
MethodInfluentialNew Sensor Data Structuring for Deeper Feature Extraction in Human Activity Recognition
Alemayoh, T. T., Lee, J. H., & Okamoto, S. (2021) · Sensors · doi:10.3390/s21082814 · Cited by 59
Larson, C., Spjut, J., Knepper, R., & Shepherd, R. (2019) · Soft Robotics · doi:10.1089/soro.2018.0086 · Cited by 51
“We initially considered other approaches to capture temporal information, including long-short term memory (LSTM) units, as well as CNN feature extractors in combination with LSTMs (Ordóñez and Roggen 2016), however we found that gestures, and even short sequences of gestures, occur over relatively short time horizons.”
MethodUser-Ranking Video Summarization With Multi-Stage Spatio–Temporal Representation
Huang, S., Li, X., Zhang, Z. M., Wu, F. Q., & Han, J. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2889265 · Cited by 58
“Although the combination of 1D-convs and LSTMs is rarely explored in computer vision community, it has been widely studied in other fields of sequential learning including text analysis [34]–[36], acoustics processing [37], and signal processing [38].”
BackgroundNegative Selection by Clustering for Contrastive Learning in Human Activity Recognition
Wang, J., Zhu, T., Chen, L. L., Ning, H., & Wan, Y. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3239945 · Cited by 52
“is fast in inference and outperforms DeepConvLSTM [30] in supervised learning.”
BackgroundApplication of ensemble RNN deep neural network to the fall detection through IoT environment
Farsi, M. A. (2020) · Alexandria Engineering Journal · doi:10.1016/j.aej.2020.06.056 · Cited by 60
A Novel Action Recognition Framework Based on Deep-Learning and Genetic Algorithms
Yılmaz, A. A., Güzel, M. S., Bostancı, E., & Askerzade, I. N. (2020) · IEEE Access · doi:10.1109/access.2020.2997962 · Cited by 60
Wang, K., He, J., & Zhang, L. (2021) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2021.3086008 · Cited by 49
“tation technique and discrete cosine transformation (DCT), then fed the 2D signal image into a two layer 2D CNN to classify its signal image equaling to the desired activity recognition. Ordez et al. [16] proposed an architecture comprised of CNN and LSTM recurrent units (DeepConvLSTM), which outperforms CNN. However,…”
BackgroundMethodHuang, W., Zhang, L., Teng, Q., Song, C., & He, J. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2021.3092396 · Cited by 56
“…78.07 [25] 97.50[25] 93.03[25] 96.90[35] 76.83[26] 74.46[35] 96.90[35] 93.50[35] Other Researchers’ Results 95.75[36] 82.30 [37] 74.66[38] 93.32[39] 93.70 [40] 95.18[12] 75.74[27] 77.27[41] 97.51[42] 86.00[43] 96.37[39] 74.50[40] - 98.20 [44] 89.96[28] stage of pre-processing is also introduced.”
BackgroundInfluentialCOVID-19 Patient Count Prediction Using LSTM
Iqbal, M. R., Al-Obeidat, F. N., Maqbool, F., Razzaq, S., Anwar, S., Tubaishat, A., Khan, M. S., & Shah, B. (2021) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2021.3056769 · Cited by 56
Muralidharan, K., Ramesh, A., Rithvik, G., Prem, S., Reghunaath, A. A., & Gopinath, M. P. (2021) · International Journal of Cognitive Computing in Engineering · doi:10.1016/j.ijcce.2021.09.001 · Cited by 69
Luo, H., Huang, M., & Zhou, Z. (2019) · Measurement · doi:10.1016/j.measurement.2019.01.038 · Cited by 58
A motion-aware ConvLSTM network for action recognition
Majd, M., & Safabakhsh, R. (2019) · Applied Intelligence · doi:10.1007/s10489-018-1395-8 · Cited by 54
Body Pose and Context Information for Driver Secondary Task Detection
Martín, M., Popp, J., Anneken, M., Voit, M., & Stiefelhagen, R. (2018) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2018.8500523 · Cited by 58
“Popular approaches use for example 3D convolutional networks applied to image stacks [19] or recurrent networks applied on image sequences [20].”
MethodFakhrulddin, A. H., Fei, X., & Li, H. (2017) · International Conference on Systems and Informatics (ICSAI) · doi:10.1109/icsai.2017.8248516 · Cited by 53
“The system that the researcher proposes in this research follows three phases such as pre-processing, characteristics mining and sorting (Ordóñez and Roggen, 2016).”
MethodDeep Dilated Convolution on Multimodality Time Series for Human Activity Recognition
Xi, R., Hou, M., Fu, M., Qu, H., & Liu, D. (2018) · 2018 International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2018.8489540 · Cited by 51
“As [19] presents, the architecture of DeepConvLSTM can be described as C(64)−C(64)−C(64)−C(64)−R(128)−R(128)−Sm.”
BackgroundConvolutional neural network and long short-term memory models for ice-jam predictions
Madaeni, F., Chokmani, K., Lhissou, R., Homayouni, S., Gauthier, Y., & Tolszczuk-Leclerc, S. (2022) · The Cryosphere · doi:10.5194/tc-16-1447-2022 · Cited by 51
DCapsNet: Deep capsule network for human activity and gait recognition with smartphone sensors
Sezavar, A., Atta, R., & Ghanbari, M. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.110054 · Cited by 61
WEAR: An Outdoor Sports Dataset for Wearable and Egocentric Activity Recognition
Bock, M., Kuehne, H., Laerhoven, K. V., & Moeller, M. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3699776 · Cited by 124
“Within the last decade deep learning based-methods have established themselves as the de facto standard in inertial-based HAR as they have shown to outperform classical machine learning algorithms [24, 25, 51].”
A Recurrent Neural Network-Based Method for Dynamic Load Identification of Beam Structures
Hongji, Y., Jiang, J., Chen, G., Mohamed, M. S., & Lü, F. (2021) · Materials · doi:10.3390/ma14247846 · Cited by 54
“Ord ó ñez et al. [46] proposed a generic deep framework for activity recognition based on convolutional and LSTM-recurrent units to capture the temporal dynamics of human activity recognition.”
BackgroundA Dataset of Routine Daily Activities in an Instrumented Home
Cumin, J., Lefebvre, G., Ramparany, F., & Crowley, J. L. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-67585-5_43 · Cited by 71
“This data is designed to be used for supervised offline activity recognition [4], supervised online activity recognition [6], unsupervised activity discovery [2], activity prediction [5], as well as other applications.”
MethodPei, L., Xia, S., Chu, L., Xiao, F., Wu, Q., Yu, W., & Qiu, R. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3055859 · Cited by 50
“Deep ConvLSTM [30]: This model uses a deep convolutional neural network to extract feature and an RNN to learn time dependencies.”
MethodInfluentialA study on IMU-Based Human Activity Recognition Using Deep Learning and Traditional Machine Learning
Hou, C. (2020) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs49078.2020.9118506 · Cited by 56
“Some researchers also had the opinions that the combinations of CNN and RNN(LSTM) may have better effects on a basis of the recent success of recurrent neural networks for time series domains in [12].”
BackgroundAkbari, A., & Jafari, R. (2020) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2019.2963816 · Cited by 50
“Several prior investigations have used data from inertial measurement units (IMU) or motion sensors to automatically extract features by leveraging convolutional neural networks (CNN) for detecting human gestures and activities [34]–[37].”
BackgroundGao, Y., Long, Y., Guan, Y., Basu, A., Baggaley, J., & Ploetz, T. (2019) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3314399 · Cited by 49
“osition, outliers, etc. 4.2 Baselines We compare the proposed DPD method to a range of alternative, conventional machine learning methods. Despite popularity, existing deep learning techniques (e.g., [22, 24, 39, 51, 52]) are not applicable to this problem because these methods need training data with sample/window-wi…”
BackgroundMethodInfluentialDFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition
Yang, Z., Raymond, O. I., Zhang, C., Wan, Y., & Long, J. (2018) · IEEE Access · doi:10.1109/access.2018.2873315 · Cited by 48
“Recently, many researchers have adopted Convolutional Neural Networks (CNNs) to deploy human activity recognition system, such as [6], [22], and [26]–[28].”
BackgroundResultInfluentialAttention-Based Residual BiLSTM Networks for Human Activity Recognition
Zhang, J., Liu, Y., & Yuan, H. (2023) · IEEE Access · doi:10.1109/access.2023.3310269 · Cited by 59
“Ordóñez and Roggen [16] proposed that the DeepConvLSTM network, which combines CNN and LSTM, is a significant improvement over single neural networks for human behavior”
BackgroundInfluentialFLAME: Federated Learning across Multi-device Environments
Cho, H., Mathur, A., & Kawsar, F. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3550289 · Cited by 79
“Previous research also investigates ways to improve the runtime system performance through sensor selection [29, 30, 32, 36, 75] and sensor fusion [51, 52, 69, 72, 73] in multi-device environments.”
BackgroundAscioglu, G., & Şenol, Y. (2020) · IEEE Access · doi:10.1109/access.2020.3024003 · Cited by 57
“Ordonez and Roggen [21] made a performance comparison between CNN and ConvLSTM using Opportunity dataset.”
MethodComprehensive machine and deep learning analysis of sensor-based human activity recognition
Balaha, H. M., & Hassan, A. E.-S. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08374-7 · Cited by 58
“In Ordonez and Roggen [66], a DL architecture using a combination of convolutional and recurrent neural networks was proposed to conduct HAR from wearable sensors.”
MethodA Deep Learning Approach to Detect Real-Time Vehicle Maneuvers Based on Smartphone Sensors
Li, P., Abdel-Aty, M. A., Cai, Q., & Islam, Z. (2020) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.3032055 · Cited by 60
Guo, L., Wang, J., Wu, Q., Li, X., Zhang, B., Zhou, L., & Xiong, D. (2023) · JMIR mHealth and uHealth · doi:10.2196/40416 · Cited by 48
Deep learning and case-based reasoning for predictive and adaptive traffic emergency management
Louati, A., Louati, H., & Li, Z. (2020) · The Journal of Supercomputing · doi:10.1007/s11227-020-03435-3 · Cited by 52
Xiang, L., Gu, Y., Gao, Z., Yu, P., Shim, V., Wang, A., & Fernandez, J. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108016 · Cited by 58
Approaching the Real-World: Supporting Activity Recognition Training with Virtual IMU Data
Kwon, H., Wang, B., Abowd, G. D., & Plötz, T. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3478096 · Cited by 48
“We compared the performance of three different classifiers: ConvLSTM [73], Random Forest with ECDF features [30], and Random Forest with classical timeseries features [10].”
MethodUnsupervised Human Activity Representation Learning with Multi-task Deep Clustering
Ma, H., Zhang, Z., Li, W., & Lu, S. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3448074 · Cited by 48
WiFi-based Human Activity Recognition using Raspberry Pi
Forbes, G. S., Massie, S., & Craw, S. (2020) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai50040.2020.00115 · Cited by 63
“The rich data streams they produce can be used to extract similar measurements to those taken during hospital testing, such as gait speed in video [13], and heart rate from wearables [14], [15].”
BackgroundKang, J., Lee, J., & Eom, D.-S. (2018) · Sensors · doi:10.3390/s18093149 · Cited by 50
“CNN and RNN models are widely used in human activity recognition using multiple sensor signals from smartphones as well [31,32].”
Jiang, Y., Zheng, L., & Ding, X. (2021) · Journal of Renewable and Sustainable Energy · doi:10.1063/5.0056980 · Cited by 44
Alqudah, A. M., & Moussavi, Z. (2025) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2025.063643 · Cited by 101
TATC: Predicting Alzheimer's Disease with Actigraphy Data
Li, J., Rong, Y., Meng, H. M. L., Lu, Z., Kwok, T. C. Y., & Cheng, H. (2018) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3219819.3219831 · Cited by 57
“Representative works include SVD [5], BOSS [23], SMTS [3], MFCC [18], and CNN [20].”
BackgroundMethodJoint Learning of Temporal Models to Handle Imbalanced Data for Human Activity Recognition
Hamad, R. A., Yang, L., Woo, W. L., & Wei, B. W. (2020) · Applied Sciences · doi:10.3390/app10155293 · Cited by 61
“The average F-score is calculated from the results of the cross-validation for all the models that has successfully been performed in [18,44,53].”
MethodSakuma, K., Abrami, A., Blumrosen, G., Lukashov, S., Narayanan, R., Ligman, J. W., Caggiano, V., & Heisig, S. (2018) · Scientific Reports · doi:10.1038/s41598-018-36834-x · Cited by 51
Embedding Recurrent Neural Networks in Wearable Systems for Real-Time Fall Detection
Torti, E., Fontanella, A., Musci, M., Blago, N., Pau, D., Leporati, F., & Piastra, M. (2019) · Microprocessors and Microsystems · doi:10.1016/j.micpro.2019.102895 · Cited by 47
Driver Identification Based on Vehicle Telematics Data using LSTM-Recurrent Neural Network
Girma, A., Yan, X., & Homaifar, A. (2019) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai.2019.00127 · Cited by 58
“This memorizing capability of LSTM has shown the best performance in many time-series tasks such as activity recognition, video captioning, language translation [13], [14].”
BackgroundAn Efficient ResNetSE Architecture for Smoking Activity Recognition from Smartwatch
Hnoohom, N., Mekruksavanich, S., & Jitpattanakul, A. (2022) · Intelligent Automation & Soft Computing · doi:10.32604/iasc.2023.028290 · Cited by 46
IF-ConvTransformer: A Framework for Human Activity Recognition Using IMU Fusion and ConvTransformer
Zhang, Y., Wang, L., Chen, H., Tian, A., Zhou, S., & Guo, Y. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3534584 · Cited by 46
“83.39% 92.98% 89.77% 89.87% 74.96% 77.16% 89.57% 89.71% 83.99% 93.05% ± 1.09% ± 0.43% ± 0.57% ± 0.29% ± 0.82% ± 0.71% ± 0.63% ± 0.55% ± 0.35% ± 0.19% is still unsatisfactory, which is a bit worse than the baseline (DeepConvLSTM).”
BackgroundMethodInfluentialWearables-based multi-task gait and activity segmentation using recurrent neural networks
Martindale, C. F., Christlein, V., Klumpp, P., & Eskofier, B. M. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.08.079 · Cited by 51
Recent Advances in Pedestrian Navigation Activity Recognition: A Review
Wang, Q., Luo, H., Wang, Q., Sun, L., Ma, Z., Zhang, C., Fu, M., & Zhao, F. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3153610 · Cited by 50
Human activity recognition: A comprehensive review
Kaur, H., Rani, V., & Kumar, M. (2024) · Expert Systems · doi:10.1111/exsy.13680 · Cited by 67
Khowaja, S. A., Yahya, B. N., & Lee, S. (2020) · Human-centric Computing and Information Sciences · doi:10.1186/s13673-020-00240-y · Cited by 54
Learning with Type-2 Fuzzy activation functions to improve the performance of Deep Neural Networks
Beke, A., & Kumbasar, T. (2019) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2019.06.016 · Cited by 50
Wu, Z., Xu, Y., Yang, Y., Zhang, C., Zhu, X., & Ji, Y. (2017) · Sensors · doi:10.3390/s17020403 · Cited by 47
Recognizing Physical Activity of Older People from Wearable Sensors and Inconsistent Data
Papagiannaki, A., Zacharaki, E. I., Kalouris, G., Kalogiannis, S., Deltouzos, K., Ellul, J., & Megalooikonomou, V. (2019) · Sensors · doi:10.3390/s19040880 · Cited by 44
Deep learning recognition of diseased and normal cell representation
Iqbal, M. S., Ahmad, I. A., Bin, L., Khan, S., & Rodrigues, J. J. P. C. (2020) · Transactions on Emerging Telecommunications Technologies · doi:10.1002/ett.4017 · Cited by 48
A Survey on Human Activity Recognition Using Deep Learning Techniques and Wearable Sensor Data
Dua, N., Singh, S. N., Challa, S. K., Semwal, V. B., & Kumar, M. L. S. S. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-24352-3_5 · Cited by 57
Towards Human Activity Recognition: A Hierarchical Feature Selection Framework
Wang, A., Chen, G., Wu, X. L., Liu, L., An, N., & Chang, C. (2018) · Sensors · doi:10.3390/s18113629 · Cited by 49
A CNN-LSTM neural network for recognition of puffing in smoking episodes using wearable sensors
Senyurek, V. Y., Imtiaz, M. H., Belsare, P. V., Tiffany, S. T., & Sazonov, E. E. (2020) · Biomedical Engineering Letters · doi:10.1007/s13534-020-00147-8 · Cited by 47
Accurate detection of sitting posture activities in a secure IoT based assisted living environment
Tariq, M. H., Majeed, H., Beg, M. O., Khan, F. A., & Derhab, A. (2018) · Future Generation Computer Systems · doi:10.1016/j.future.2018.02.013 · Cited by 55
Thakur, D., Dangi, S., & Lalwani, P. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106870 · Cited by 51
Cross-Domain HAR: Few Shot Transfer Learning for Human Activity Recognition
Thukral, M., Haresamudram, H., & Ploetz, T. (2023) · arXiv · doi:10.48550/arxiv.2310.14390 · Cited by 45
Deep Learning for Sensor-Based Rehabilitation Exercise Recognition and Evaluation
Zhu, Z.-A., Lu, Y.-C., You, C.-H., & Chiang, C.-K. (2019) · Sensors · doi:10.3390/s19040887 · Cited by 49
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.