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 901–1,000 of 2,807
Learning to judge like a human: convolutional networks for classification of ski jumping errors
Brock, H., Ohgi, Y., & Lee, J. (2017) · ACM International Symposium on Wearable Computers (ISWC) · doi:10.1145/3123021.3123038 · Cited by 23
“Instead, they apply one-dimensional filter along the temporal axis only [12, 20, 25, 26].”
BackgroundInfluentialhyper-sinh: An accurate and reliable function from shallow to deep learning in TensorFlow and Keras
Parisi, L., Ma, R., RaviChandran, N., & Lanzillotta, M. (2021) · Machine Learning with Applications · doi:10.1016/j.mlwa.2021.100112 · Cited by 43
On the need of machine learning as a service for the internet of things
Bacciu, D., Chessa, S., Gallicchio, C., & Micheli, A. (2017) · Proceedings of the 1st International Conference on Internet of Things and Machine Learning · doi:10.1145/3109761.3109783 · Cited by 20
“…activities [1, 3, 4, 22, 28, 53] identification of human patterns of mobility in indoor environments [7, 16, 39] human activity recognition [5, 23, 56, 58, 64, 71, 74] human gestures recognition [22, 23, 36, 41, 54, 59] recognition of sports activities [22, 62] estimation of human indoor…”
BackgroundXia, S., Chu, L., Pei, L., Yu, W., & Qiu, R. C. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3173957 · Cited by 20
“2) DeepConvLSTM [21]: A hybrid network with the convolutional layers and LSTM layers for the feature extraction.”
BackgroundMethodSelf-Attention-Based Deep Convolution LSTM Framework for Sensor-Based Badminton Activity Recognition
Deng, J., Zhang, S., & Ma, J. (2023) · Sensors · doi:10.3390/s23208373 · Cited by 27
“[17] 94.30% 93.98% DeepConvLSTM_Att [18] 97.34% 97.15% DeepSense [9] 95.55% 95.18% AttnSense [10] 94.23% 93.86% SelfAttnNet [27] 95 It can be seen from Table 1 that our SADeepConvLSTM framework achieves the best recognition result compared to other methods, with the average recognition accuracy…”
BackgroundMethodResultInfluentialYousefpour, N., & Pouragha, M. (2022) · International Journal for Numerical and Analytical Methods in Geomechanics · doi:10.1002/nag.3372 · Cited by 20
Deep Learning Multi-Class Approach for Human Fall Detection Based on Doppler Signatures
Cárdenas, J. D., Gutierrez, C. A., & Aguilar‐Ponce, R. M. (2023) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph20021123 · Cited by 20
Dynamic Hand Gesture Recognition for Mobile Systems Using Deep LSTM
Sarkar, A., Gepperth, A., Handmann, U., & Kopinski, T. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-72038-8_3 · Cited by 24
“The authors of [5] present DeepConvLSTM, a deep architecture fusing convolutional layers and recurrent layers from an LSTM for Human Activity Recognition (HAR).”
BackgroundDas, T., Gohain, L., Kakoty, N. M., Malarvili, M. B., Widiyanti, P., & Kumar, G. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.01.061 · Cited by 23
Zhou, B., Suh, S., Rey, V. F., Altamirano, C. A. V., & Lukowicz, P. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3534610 · Cited by 28
“In the broader activity recognition field, combinations of convolutional layers for the sensor channels and long short-term memory (LSTM) layers for the temporal domain were evaluated for motion sensors [64].”
BackgroundMethodSiamese Networks for Weakly Supervised Human Activity Recognition
Sheng, T., & Huber, M. (2019) · 2019 IEEE International Conference on Systems, Man and Cybernetics (SMC) · doi:10.1109/smc.2019.8914045 · Cited by 21
“been developed and applied in HAR systems [2], [3], [4].”
MethodEstimation of Blood Alcohol Concentration From Smartphone Gait Data Using Neural Networks
Li, R., Balakrishnan, G. P., Nie, J., Li, Y., Agu, E., Grimone, K., Herman, D. S., Abrantes, A. M., & Stein, M. D. (2021) · IEEE Access · doi:10.1109/access.2021.3054515 · Cited by 24
“ConvLSTM [8] and DeepSense [29] are the state of the art HAR architectures.”
BackgroundLarge Language Models Are Zero-Shot Recognizers for Activities of Daily Living
Civitarese, G., Fiori, M., Choudhary, P., & Bettini, C. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3725856 · Cited by 67
“• DeepConvLSTM [40]: Also known as Deep Convolutional and LSTM Recurrent Neural Network .”
MV-Sports: A Motion and Vision Sensor Integration-Based Sports Analysis System
Zhang, C., Yang, F., Li, G., Zhai, Q., Jiang, Y. F., & Xuan, D. (2018) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom.2018.8485910 · Cited by 22
“DeepConvLSTM [21] presents a deep learning framework for”
BackgroundYaagoubi, R., Lakber, C.-E., & Miky, Y. H. (2024) · Journal of Land Use Science · doi:10.1080/1747423x.2024.2403789 · Cited by 20
“The computation of the hidden value h t of an LSTM cell is updated at every time step t as follows (Ordóñez & Roggen, 2016): where i , f , o and c are, respectively, the input gate, forget gate, output gate and cell activation vectors, all of which are the same size as vector h defining the hidden…”
Change-point detection with deep learning: A review
Xu, R., Song, Z., Wu, J., Wang, C., & Zhou, S. (2025) · Frontiers of Engineering Management · doi:10.1007/s42524-025-4109-z · Cited by 33
“The extracted features are then passed to RNNs, which model the temporal dependencies across the sequence and classify activities (Ordóñez and Roggen 2016; Thodoroff et al., 2016; Chambers and Yoder 2020; Xia et al., 2020).”
Xiang, L., Gu, Y., Deng, K., Gao, Z., Shim, V., Wang, A., & Fernandez, J. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01677-0 · Cited by 22
“In this context, one-dimensional convolutional neural network (CNN1D) and long short-term memory (LSTM) architectures have demonstrated strong capabilities in predicting time-series data for human activity recognition 32,33 , gait analysis 30,34 , and running biomechanics 24,35 .”
Human Activity Recognition for Indoor Localization Using Smartphone Inertial Sensors
Moreira, D., Barandas, M., Rocha, T., Alves, P., Santos, R., Leonardo, R., Vieira, P. A., & Gamboa, H. F. S. (2021) · Sensors · doi:10.3390/s21186316 · Cited by 28
On specialized window lengths and detector based human activity recognition
Li, H., Abowd, G. D., & Plötz, T. (2018) · Proceedings of the 2018 ACM International Symposium on Wearable Computers · doi:10.1145/3267242.3267246 · Cited by 19
“In contrast to the periodic activities covered by most of the other datasets, activities in Opportunity are largely sporadic and vary a lot (in duration, and general appearance) within classes [8, 19].”
BackgroundMethodMulti-modal egocentric activity recognition using multi-kernel learning
Arabacı, M. A., Özkan, F., Sürer, E., Jančovič, P., & Temizel, A. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-08789-7 · Cited by 22
“For that purpose, sensor streams are converted by using channel [20] or model [43] approaches to fit into deep learning algorithms.”
BackgroundMethodChen, H., Xiong, F., Wu, D., Zheng, L., Peng, A., Hong, X., Tang, B., Lü, H., Shi, H., & Zheng, H. (2017) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm.2017.8217821 · Cited by 23
Continual Activity Recognition with Generative Adversarial Networks
Ye, J., Nakwijit, P., Schiemer, M., Jha, S., & Zambonelli, F. (2021) · ACM Transactions on Internet of Things · doi:10.1145/3440036 · Cited by 23
“The classifier for activity recognition is implemented as a fully-connected neural network with two hidden layers, which is based on the recent studies on similar types of data [42], [56], [35].”
MethodZhu, Y., Luo, H., Chen, R., & Zhao, F. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3285547 · Cited by 25
“There also exist plenty of works to jointly learn spatial–temporal features using CNN-RNN deep models [20], [21] for HAR.”
BackgroundMethodMukherjee, P., & Mukherjee, A. (2019) · Elsevier eBooks · doi:10.1016/b978-0-12-819361-7.00001-4 · Cited by 25
Ding, W., Abdel‐Basset, M., Hawash, H., & Moustafa, N. (2022) · Information Sciences · doi:10.1016/j.ins.2022.03.046 · Cited by 20
Impact of Wireless Sensor Data Mining with Hybrid Deep Learning for Human Activity Recognition
Nair, R., Ragab, M., Mujallid, O. A., Mohammad, K. A., Mansour, R. F., & Viju, G. K. (2022) · Wireless Communications and Mobile Computing · doi:10.1155/2022/9457536 · Cited by 23
“For the opportunity dataset, Ordonez and Roggen [15] implemented DeepConvLSTM with a convolutional and recurrent layer combination to achieve 95.8%, whereas Hammerla et al. [16] used bidirectional LSTM with two parallel recurrent layers that reach into the “future” and the “past.””
MethodInfluentialAn eXplainable Self-Attention-Based Spatial–Temporal Analysis for Human Activity Recognition
Meena, T., & Sarawadekar, K. P. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3335449 · Cited by 24
“Furthermore, global average pooling layer is implemented for parameter reduction [21] thus replacing the fully connected layer in [20].”
MethodSenseCrypt: A Security Framework for Mobile Crowd Sensing Applications
Owoh, N. P., & Singh, M. M. (2020) · Sensors · doi:10.3390/s20113280 · Cited by 22
Block-Wise Training Residual Networks on Multi-Channel Time Series for Human Activity Recognition
Teng, Q., Zhang, L., Tang, Y., Song, S., Wang, X., & He, J. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3085360 · Cited by 21
Liu, C.-P., Li, J.-H., Chu, E.-P., Hsieh, C.-Y., Liu, K.-C., Chan, C.-T., & Tsao, Y. (2023) · IEEE International Symposium on Medical Measurements and Applications (MeMeA) · doi:10.1109/memea57477.2023.10171944 · Cited by 20
“[14] has been widely used to compare FD tasks [14, 16, 17].”
MethodHead-AR: Human Activity Recognition with Head-Mounted IMU Using Weighted Ensemble Learning
Gjoreski, H., Kiprijanovska, I., Stankoski, S., Kalabakov, S., Broulidakis, J., Nduka, C. C., & Gjoreski, M. (2020) · Smart innovation, systems and technologies · doi:10.1007/978-981-15-8944-7_10 · Cited by 21
Learning-Based Complex Motion Patterns Recognition for Pedestrian Dead Reckoning
Luo, Y., Guo, C., Su, J., Guo, W., & Zhang, Q. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3029719 · Cited by 24
“What’s more, it has shown that the neural network is superior to traditional machine learning methods [14], [15].”
BackgroundMethodMicro-activity recognition in industrial assembly process with IMU data and deep learning
Sopidis, G., Haslgrübler, M., Azadi, B., Anzengruber-Tánase, B., Ahmad, A., Ferscha, A., & Baresch, M. (2022) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3529190.3529204 · Cited by 23
“Ordonez et al. [Ordóñez and Roggen 2016] proposed a combination of convolutional and LSTM recurrent units, which is appropriate for multi-modal wearable sensors and eliminates the requirement for expertise in designing features.”
BackgroundMekruksavanich, S., & Jitpattanakul, A. (2024) · Applied System Innovation · doi:10.3390/asi7040059 · Cited by 22
“In order to emphasize the originality and effectiveness of our CNN-ResBiGRU model, we performed a comparative assessment of its performance against other cutting-edge models often used for time-series classification tasks, including InceptionTime [42], Deep-ConvLSTM [43], and ResNet [44].”
MethodOlder Adult Fall Risk Prediction with Deep Learning and Timed Up and Go (TUG) Test Data
Maiora, J., Rezola‐Pardo, C., García, G., Sanz, B., & Graña, M. (2024) · Bioengineering · doi:10.3390/bioengineering11101000 · Cited by 30
“Two of the most popular deep learning networks are the convolutional neural network (CNN) [41] and the long short-term memory (LSTM) [42].”
BackgroundMethodA Hierarchical Approach in Food and Drink Intake Recognition Using Wearable Inertial Sensors
Anderez, D. O., Lotfi, A., & Langensiepen, C. S. (2018) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3197768.3201542 · Cited by 22
“Deep learning techniques are increasingly in human activity recognition systems [12, 18].”
BackgroundHaque, S. T., Debnath, M., Yasmin, A., Mahmud, T., & Ngu, A. H. H. (2024) · Sensors · doi:10.3390/s24196235 · Cited by 25
“In [21], a DeepConvLSTM was presented for recognizing two families of human activities (periodic motion activities, such as walking and bicycling, and gestures such as drinking water from a cup) on the Skoda and OPPORTUNITY datasets.”
MethodDeep Neural Network Sleep Scoring Using Combined Motion and Heart Rate Variability Data
Haghayegh, S., Khoshnevis, S., Smolensky, M. H., Diller, K. R., & Castriotta, R. J. (2020) · Sensors · doi:10.3390/s21010025 · Cited by 25
Deep fusion of heterogeneous sensor data
Liu, Z., Zhang, W., Quek, T. Q. S., & Lin, S. (2017) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2017.7953301 · Cited by 21
Robust Feature Representation Using Multi-Task Learning for Human Activity Recognition
Azadi, B., Haslgrübler, M., Anzengruber-Tánase, B., Sopidis, G., & Ferscha, A. (2024) · Sensors · doi:10.3390/s24020681 · Cited by 26
“In their study, a symmetric auto-encoder, inspired by [41], is employed as the unsupervised feature extractor.”
MethodMulti-Class Wheat Moisture Detection with 5GHz Wi-Fi: A Deep LSTM Approach
Yang, W., Wang, X., Cao, S., Wang, H., & Mao, S. (2018) · International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn.2018.8487423 · Cited by 19
“In fact, the deep LSTM network has a stronger data representation capability than traditional machine learning methods [32], [33], which has been successfully applied for speech recognition [34], human activity recognition [35], and indoor localization [36].”
BackgroundThukral, M., Dhekane, S. G., Hiremath, S. K., Haresamudram, H., & Plötz, T. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3712278 · Cited by 52
Multi-modal lifelog data fusion for improved human activity recognition: A hybrid approach
Oh, Y., & Kim, S. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102464 · Cited by 25
Baghezza, R., Bouchard, K., Bouzouane, A., & Gouin-Vallerand, C. (2021) · Sensors · doi:10.3390/s21082786 · Cited by 25
Conformer-Based Human Activity Recognition Using Inertial Measurement Units
Seenath, S., & Dharmaraj, M. (2023) · Sensors · doi:10.3390/s23177357 · Cited by 22
Recognition of Fine-Grained Walking Patterns Using a Smartwatch with Deep Attentive Neural Networks
Kim, H., Kim, H.-J., Park, J., Ryu, J.-K., & Kim, S.-C. (2021) · Sensors · doi:10.3390/s21196393 · Cited by 20
“For example, an earlier study investigated the possibility of using a smartwatch to recognize 18 different everyday activities [20].”
BackgroundTemporal Feature Alignment in Contrastive Self-Supervised Learning for Human Activity Recognition
Khaertdinov, B., & Asteriadis, S. (2022) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb54206.2022.10007984 · Cited by 28
No Interface, No Problem: Gesture Recognition on Physical Objects Using Radar Sensing
Attygalle, N. T., Leiva, L. A., Kljun, M., Sandor, C., Plopski, A., Kato, H., & Pucihar, K. Č. (2021) · Sensors · doi:10.3390/s21175771 · Cited by 20
“The hybrid model, depicted in Figure 2, is a deep CNN+LSTM architecture inspired by previous work [47,48,53,55–57].”
MethodLearning to rank: An intelligent system for person reidentification
Khan, S. U., Haq, I. U., Khan, N., Muhammad, K., Hijji, M., & Baik, S. W. (2022) · International Journal of Intelligent Systems · doi:10.1002/int.22820 · Cited by 19
“In a conventional CNN, the convolutional filters give a smaller receptive field after applying the pooling layer operation.(35) To decrease the likelihood of overfitting, a large number of pooling layers are used in a deep learning model, which often causes a loss of significant information.”
MethodA Framework For Mode-Free Prosthetic Control For Unstructured Terrains
Rai, V., & Rombokas, E. (2019) · International Conference on Rehabilitation Robotics · doi:10.1109/icorr.2019.8779439 · Cited by 20
“Most HAR systems are trained on chunks of time-series and have difficulty ”scaling up” to recognize complex highlevel behavior which can have hour-long to day-long duration [12].”
BackgroundBo, Z., Li, H., Yang, H., Li, C., Wu, S., Xu, C., Xiong, G., Mariotti, D., Yan, J., Cen, K., et al. (2021) · International Journal of Heat and Mass Transfer · doi:10.1016/j.ijheatmasstransfer.2021.121075 · Cited by 19
Meena, T., & Sarawadekar, K. P. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3301187 · Cited by 21
“are primarily two methods for labeling sliding windows [13].”
MethodMulticlass autoencoder-based active learning for sensor-based human activity recognition
Park, H., Lee, G. H., Han, J., & Choi, J. K. (2023) · Future Generation Computer Systems · doi:10.1016/j.future.2023.09.029 · Cited by 22
Cui, H., & Chang, C. (2020) · IEEE Access · doi:10.1109/access.2020.3003652 · Cited by 19
“Advanced human motion-modeling mainly involves four types of spe-cific learning systems and their application for movement generation, according to DNN’s topological structure [21].”
MethodALGAN: Time Series Anomaly Detection with Adjusted-LSTM GAN
Bashar, M. A., & Nayak, R. (2025) · International Journal of Data Science and Analytics · doi:10.1007/s41060-025-00810-2 · Cited by 40
Zhang, Y., & Gu, D. (2019) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp-bmei48845.2019.8965723 · Cited by 22
Quantum-Inspired Support Vector Machines for Human Activity Recognition in Industry 4.0
Agarwal, P., & Alam, M. (2022) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-16-6289-8_24 · Cited by 22
A MEMS Gyroscope Noise Suppressing Method Using Neural Architecture Search Neural Network
Zhu, Z., Bo, Y., & Jiang, C. (2019) · Mathematical Problems in Engineering · doi:10.1155/2019/5491243 · Cited by 18
Improving Inertial Sensor-Based Activity Recognition in Neurological Populations
Çelik, Y., Aslan, M. F., Sabancı, K., Stuart, S., Woo, W. L., & Godfrey, A. (2022) · Sensors · doi:10.3390/s22249891 · Cited by 21
Quantifying mating behaviour using accelerometry and machine learning: challenges and opportunities
Aulsebrook, A. E., Jacques-Hamilton, R., & Kempenaers, B. (2023) · Animal Behaviour · doi:10.1016/j.anbehav.2023.10.013 · Cited by 18
Improving Cross-Subject Activity Recognition via Adversarial Learning
Leite, C. F. S., & Xiao, Y. (2020) · IEEE Access · doi:10.1109/access.2020.2993818 · Cited by 18
“The performances for DeepConvLSTM [26] and the LSTM with Uniqueness Attention [41] are also reported.”
BackgroundMethodInfluentialAmin, M., Ullah, K., Asif, M., Shah, H., Mehmood, A., & Khan, M. A. (2023) · Diagnostics · doi:10.3390/diagnostics13111897 · Cited by 19
Children Activity Recognition: Challenges and Strategies
Hosseini, A., Fazeli, S., van Vliet, E., Valencia, L., Habre, R., Sarrafzadeh, M., & Bui, A. A. T. (2018) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2018.8513320 · Cited by 17
“Ordez et al. [5] then compared recurrent neural networks (RNN) with other deep learning models and found that RNNs outperformed non-recurrent neural networks by an average of 4%.”
BackgroundMohamed, A. S., Hassan, N. F., & Jamil, A. S. (2024) · Cybernetics and Information Technologies · doi:10.2478/cait-2024-0031 · Cited by 26
Chen, L., & Hu, D. (2022) · Advanced Robotics · doi:10.1080/01691864.2022.2160274 · Cited by 24
“LSTM is brought up to deal with sequential information with the help of memory cells, which store and output information aiming at learning long time scales temporal relationships [48,50].”
BackgroundDeep Learning Forecasting Model for Market Demand of Electric Vehicles
Şimşek, A. İ., Koç, E., Taşdemir, B. D., Aksöz, A., Türkoğlu, M., & Şengür, A. (2024) · Applied Sciences · doi:10.3390/app142310974 · Cited by 26
Taleb, C., Khachab, M. Y., Mokbel, C., & Likforman-Sulem, L. (2019) · International Conference on Document Analysis and Recognition Workshops (ICDARW) · doi:10.1109/icdarw.2019.50111 · Cited by 20
“In addition, the BLSTM-based model takes advantage of learning the temporal feature activation dynamics, which the CNN model is not capable to model [14].”
MethodJaveed, M., Mudawi, N. A., Alabduallah, B. I., Jalal, A., & Kim, W. (2023) · Sensors · doi:10.3390/s23104716 · Cited by 18
“The following subsections will explain the three techniques in detail.”
Deep reinforcement learning-aided autonomous navigation with landmark generators
Wang, X., Sun, Y., Xie, Y., Bin, J., & Xiao, J. H. (2023) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2023.1200214 · Cited by 26
STranGAN: Adversarially-learnt Spatial Transformer for scalable human activity recognition
Faridee, A. Z. M., Chakma, A., Misra, A., & Roy, N. (2021) · Smart Health · doi:10.1016/j.smhl.2021.100226 · Cited by 21
Deep Crash Detection From Vehicular Sensor Data With Multimodal Self-Supervision
Kubin, L., Bianconcini, T., de Andrade, D. C., Simoncini, M., Taccari, L., & Sambo, F. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3114816 · Cited by 19
The Wearable Radar: Sensing Gestures Through Fabrics
Leiva, L. A., Kljun, M., Sandor, C., & Pucihar, K. Č. (2020) · International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI) · doi:10.1145/3406324.3410720 · Cited by 19
“We built a hybrid deep CNN+LSTM (convolutional neural network + long short-term memory) model, inspired by previous work on human activity recognition [5, 6, 8, 9], which is depicted in Figure 4.”
MethodTowards application of non-invasive environmental sensors for risks and activity detection
Dimitrievski, A., Zdravevski, E., Lameski, P., & Trajkovik, V. (2016) · IEEE International Conference on Intelligent Computer Communication and Processing (ICCP) · doi:10.1109/iccp.2016.7737117 · Cited by 21
“We expect that sensor fusion using bigger number and variety of sensors will enable better feature representation.”
BackgroundZhang, P., & Zhang, J. (2021) · Robotica · doi:10.1017/s0263574721000801 · Cited by 19
“Ordóñez [25] proposed a general depth framework for activity recognition based on convolutional neural network (CNN) and LSTM, which fuses multimodal sensors to improve performance.”
BackgroundMachine Learning Applied to Datasets of Human Activity Recognition: Data Analysis in Health Care
Ariza-Colpas, P., Enrico, V., Butt, S. A., De-La-Hoz-Franco, E., Alberto, P.-M. M., Carrascal, A. I. O., Tariq, M. I., Restrepo, J. K. G., & Patara, F. (2022) · Current Medical Imaging Formerly Current Medical Imaging Reviews · doi:10.2174/1573405618666220104114814 · Cited by 21
Attribute Representation for Human Activity Recognition of Manual Order Picking Activities
Reining, C., Schlangen, M., Hissmann, L., ten Hompel, M., Rueda, F. M., & Fink, G. A. (2018) · International Workshop on Sensor-based Activity Recognition and Interaction (iWOAR) · doi:10.1145/3266157.3266214 · Cited by 22
“For both datasets, attributes [ a 1 , 2 , 3 , a 9 , a 13 , 14 , 15 ] yield the best performances.”
BackgroundMethodInfluentialPham, C.-H., Huynh‐The, T., Gooya, E. S., Elbouz, M., & Alfalou, A. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107854 · Cited by 24
Beyond Mobile Apps: A Survey of Technologies for Mental Well-Being
Woodward, K., Kanjo, E., Brown, D. J., McGinnity, T. M., Inkster, B., Macintyre, D. J., & Tsanas, A. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2020.3015018 · Cited by 18
“Overall, tangible interfaces and apps provide new opportunities to enhance existing assessment methods, as the convenience and additional functionality lead these technological alternatives to improve the reporting and communicating of mental well-being.”
BackgroundA Hybrid Deep Representation Learning Model for Time Series Classification and Prediction
Guo, Y., Wu, Z., & Ji, Y. (2017) · International Conference on Big Data Computing and Communications (BigCom) · doi:10.1109/bigcom.2017.13 · Cited by 18
Zhao, Y., Guo, S., Chen, Z., Shen, Q., Meng, Z., & Xu, H. (2022) · Applied Sciences · doi:10.3390/app12115408 · Cited by 21
Temporal Approaches for Human Activity Recognition Using Inertial Sensors
Garcia, F. A., Ranieri, C. M., & Romero, R. A. F. (2019) · Latin American Robotics Symposium, Brazilian Symposium on Robotics and Workshop on Robotics in Education (LARS/SBR/WRE) · doi:10.1109/lars-sbr-wre48964.2019.00029 · Cited by 21
High-Performance Real-Time Human Activity Recognition Using Machine Learning
Pardhu, T., Acharya, B., & Moreira, F. (2024) · Mathematics · doi:10.3390/math12223622 · Cited by 26
Intelligent Fault Diagnosis of Liquid Rocket Engine via Interpretable LSTM with Multisensory Data
Zhang, X., Hua, X., Zhu, J., & Ma, M. (2023) · Sensors · doi:10.3390/s23125636 · Cited by 22
A Large-Scale Study in Predictability of Daily Activities and Places
Moon, G. E., & Hamm, J. (2016) · Mobile Computing, Applications, and Services · doi:10.4108/eai.30-11-2016.2267041 · Cited by 19
Deep Convolutional Network with Long Short-Term Memory Layers for Dynamic Gesture Recognition
Siriak, R., Skarga-Bandurova, I., & Boltov, Y. (2019) · IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) · doi:10.1109/idaacs.2019.8924381 · Cited by 22
Machine learning for intraoperative prediction of viability in ischemic small intestine
Strand-Amundsen, R. J., Tronstad, C., Reims, H. M., Reinholt, F. P., Høgetveit, J. O., & Tønnessen, T. I. (2018) · Physiological Measurement · doi:10.1088/1361-6579/aae0ea · Cited by 18
“This type of network is suitable for multimodal wearable sensors, can perform sensor fusion naturally, does not require expert knowledge in designing features and explicitly models the temporal dynamics of feature activations (Ordonez and Roggen 2016).”
BackgroundJamil, H., Qayyum, F., Iqbal, N., Jamil, F., & Kim, D. H. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3228120 · Cited by 22
A Temporal Order Modeling Approach to Human Action Recognition from Multimodal Sensor Data
Ye, J., Hu, H., Qi, G.-J., & Hua, K. A. (2017) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3038917 · Cited by 18
“DOI: http://dx.doi.org/10.1145/3038917”
InfluentialNguyen, V. S., Kim, H.-S., & Suh, D. (2023) · IEEE Access · doi:10.1109/access.2023.3338137 · Cited by 17
“However, due to the inherent variability in data collection (where 24 participants performed various actions, introducing substantial noise and variance), achieving high classification accuracy was challenging, as similarly observed in previous studies [20], [48].”
MethodResultInfluentialAugToAct: scaling complex human activity recognition with few labels
Faridee, A. Z. M., Khan, M. A. A. H., Pathak, N., & Roy, N. (2019) · Proceedings of the 16th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services · doi:10.1145/3360774.3360831 · Cited by 17
Umbricht, D., Cheng, W., Lipsmeier, F., Bamdadian, A., & Lindemann, M. (2020) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2020.574375 · Cited by 27
Multisensor-Based 3D Gesture Recognition for a Decision-Making Training System
Pan, T., Chang, C.-Y., Tsai, W.-L., & Hu, M. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3012887 · Cited by 21
“For instance, Ordóñez and Roggen [29] proposed a framework composed of convolution layers and Long Short-Term Memory Network (LSTM) recurrent layers, which has the more satisfied results for learning the feature representations and temporal dependencies necessary than only consider convolutional…”
BackgroundClassification of Motion Sickness Levels using Multimodal Biosignals in Real Driving Conditions
Hwang, J.-U., Bang, J.-S., & Lee, S. (2022) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc53654.2022.9945559 · Cited by 23
“The consecutive application of 1DCNN and LSTM was proposed for the popularity of 1DCNN in the field of sequential data analysis [33] and the effectiveness of CNN-LSTM in extracting features and their temporal relations in sequential data [34].”
MethodAutonomous Human Activity Classification From Wearable Multi-Modal Sensors
Lu, Y., & Velipasalar, S. (2019) · IEEE Sensors Journal · doi:10.1109/jsen.2019.2934678 · Cited by 21
“Ordóñez and Roggen [8] use accelerometer and gyroscope data, and employ convolutional and Long Short Term Memory (LSTM) recurrent units for activity recognition.”
MethodAmerineni, R., Gupta, L., Steadman, N., Annauth, K., Burr, C., Wilson, S., Barnaghi, P. M., & Vaidyanathan, R. (2021) · Sensors · doi:10.3390/s21248409 · Cited by 17
Chai, S., Jain, R. K., Liu, J., Teng, S., Tateyama, T., Li, Y., & Chen, Y. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127482 · Cited by 25
Automatic Annotation of Unlabeled Data from Smartphone-Based Motion and Location Sensors
Owoh, N. P., Singh, M. M., & Zaaba, Z. F. (2018) · Sensors · doi:10.3390/s18072134 · Cited by 21
Zhou, Y., King, T., Zhao, H., Huang, Y., Riedel, T., & Beigl, M. (2024) · Proceedings of the 2024 ACM International Symposium on Wearable Computers · doi:10.1145/3675095.3676624 · Cited by 17
“Variants of RNNs, such as Gated Recurrent Units (GRU) [29], Long Short-Term Memory (LSTM) [25], and bidirectional-LSTM [38], are commonly used.”
BackgroundMethodInfluentialUncovering Human Multimodal Activity Recognition with a Deep Learning Approach
Ranieri, C. M., Vargas, P. A., & Romero, R. A. F. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9207255 · Cited by 21
“Most advances on video classification were built on the Two-Stream ConvNet [4], whereas satisfactory results on inertial data have been provided by the combination of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) [5].”
MethodAttention-based LSTM Network for Wearable Human Activity Recognition
Sun, B., Liu, M., Zheng, R., & Zhang, S. (2019) · Chinese Control Conference (CCC) · doi:10.23919/chicc.2019.8865360 · Cited by 20
“Some simple activity-aware systems are now commercial in the form of fitness tracker or fall detection devices [4].”
Background
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.