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 201–300 of 2,807
Robust Human Activity Recognition using smartwatches and smartphones
San-Segundo, R., Blunck, H., Moreno-Pimentel, J., Stisen, A., & Gil-Martín, M. (2018) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2018.04.002 · Cited by 105
Yao, R., Lin, G., Shi, Q., & Ranasinghe, D. C. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2017.12.024 · Cited by 105
Assessing the State of Self-Supervised Human Activity Recognition Using Wearables
Haresamudram, H., Essa, I., & Plötz, T. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3550299 · Cited by 107
Shoulder physiotherapy exercise recognition: machine learning the inertial signals from a smartwatch
Burns, D. M., Leung, N., Hardisty, M., Whyne, C. M., Henry, P., & McLachlin, S. (2018) · Physiological Measurement · doi:10.1088/1361-6579/aacfd9 · Cited by 103
“…convolutional input layers to generate a feature representation of the signal tensor which is learned by the recurrent layers utilizing long short term memory (LSTM) cells, an approach utilized recently for video (Yue-Hei Ng et al. 2015) and inertial signal (Ordonez and Roggen 2016) classification.”
BackgroundMethodDeep Multimodal Representation Learning from Temporal Data
Yang, X., Ramesh, P., Chitta, R., Madhvanath, S., Bernal, E. A., & Luo, J. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.538 · Cited by 104
“…the contribution of different aspects of the CorrRNN design, we also evaluate different model configurations summarized in hidden layer representations obtained from each network are then reduced to 256 dimensions using PCA, and concatenated to obtain a 512 -dimensional fused representation.”
MethodCasilari, E., Lora-Rivera, R., & García-Lagos, F. (2020) · Sensors · doi:10.3390/s20051466 · Cited by 113
deepGesture: Deep learning-based gesture recognition scheme using motion sensors
Kim, J., Hong, G.-S., Kim, B., & Dogra, D. P. (2018) · Displays · doi:10.1016/j.displa.2018.08.001 · Cited by 112
Zebin, T., Scully, P., Peek, N. B., Casson, A. J., & Ozanyan, K. B. (2019) · IEEE Access · doi:10.1109/access.2019.2941836 · Cited by 119
“[15] R. Chavarriaga, H. Sagha, A. Calatroni, S. T. Digumarti, G. Tröster, J. del R. Millán, and D. Roggen, ‘‘The Opportunity challenge: A benchmark database for on-body sensor-based activity recognition,’’ Pattern Recognit.”
MethodInfluentialNormal and pathological gait classification LSTM model
Khokhlova, M., Migniot, C., Morozov, A. A., Sushkova, O. S., & Dipandà, A. (2019) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2018.12.007 · Cited by 106
Skeleton-based human activity recognition using ConvLSTM and guided feature learning
Yadav, S. K., Tiwari, K., Pandey, H. M., & Akbar, S. (2021) · Soft Computing · doi:10.1007/s00500-021-06238-7 · Cited by 108
Koşar, E., & Barshan, B. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106529 · Cited by 118
Jafari, A., Ganesan, A., Thalisetty, C. S. K., Sivasubramanian, V., Oates, T., & Mohsenin, T. (2018) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2018.2848647 · Cited by 103
“Ordóñez and Roggen [10] proposed a generic deep framework for activity recognition based on convolutional and LSTM recurrent units and evaluated their framework on the Opportunity and the Skoda datasets.”
BackgroundIMU-to-Segment Assignment and Orientation Alignment for the Lower Body Using Deep Learning
Zimmermann, T., Taetz, B., & Bleser, G. (2018) · Sensors · doi:10.3390/s18010302 · Cited by 99
“The configurations follow the general framework described in [18] with modifications described subsequently.”
BackgroundMethodResultInfluentialJointly network: a network based on CNN and RBM for gesture recognition
Cheng, W., Sun, Y., Li, G., et al. (2018) · Neural Computing and Applications · doi:10.1007/s00521-018-3775-8 · Cited by 100
A Novel Human Activity Recognition and Prediction in Smart Home Based on Interaction
Du, Y., Lim, Y., & Tan, Y. (2019) · Sensors · doi:10.3390/s19204474 · Cited by 120
“In recent years, deep learning techniques have made good progress to recognize the human posture by using the data produced by wearable devices [29,30].”
BackgroundMethodSuh, S., Rey, V. F., & Lukowicz, P. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.110143 · Cited by 107
Hu, B., Dixon, P. C., Jacobs, J. V., Dennerlein, J. T., & Schiffman, J. M. (2018) · Journal of Biomechanics · doi:10.1016/j.jbiomech.2018.01.005 · Cited by 99
Deep Learning for Human Activity Recognition in Mobile Computing
Plotz, T., & Guan, Y. (2018) · Computer · doi:10.1109/mc.2018.2381112 · Cited by 99
Human Action Recognition from Various Data Modalities: A ReviewPreprint
Rahmani, H., Bennamoun, M., & Ke, Q. (2021) · doi:10.36227/techrxiv.13708270 · Cited by 105
Antwi‐Afari, M. F., Qarout, Y., Herzallah, R., Anwer, S., Umer, W., Zhang, Y., & Manu, P. (2022) · Automation in Construction · doi:10.1016/j.autcon.2022.104181 · Cited by 118
Information fusion and artificial intelligence for smart healthcare: a bibliometric study
Chen, X., Xie, H., Li, Z., Cheng, G., Leng, M., & Wang, F. L. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2022.103113 · Cited by 99
Gjoreski, M., Janko, V., Slapničar, G., Mlakar, M., Reščič, N., Bizjak, J., Drobnič, V., Marinko, M., Mlakar, N., Luštrek, M., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.04.004 · Cited by 96
“In 2016, Ordonez et al. introduced the DeepConvLSTM [15] , an architecture that stacks CNN layers over a multimodal long short-term memory (LSTM) recurrent neural network (RNN) for AR [16] .”
BackgroundHand Gesture Recognition using Deep Feature Fusion Network based on Wearable Sensors
Yuan, G., Liu, X., Yan, Q., Qiao, S., Wang, Z., & Yuan, L. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3014276 · Cited by 130
“Ordóñez et al. [21] proposed a generic deep framework for activity recognition based on convolutional and LSTM recurrent units.”
BackgroundLARa: Creating a Dataset for Human Activity Recognition in Logistics Using Semantic Attributes
Niemann, F., Reining, C., Rueda, F. M., Nair, N. R., Steffens, J. A., Fink, G. A., & ten Hompel, M. (2020) · Sensors · doi:10.3390/s20154083 · Cited by 120
“They combine learnable convolutional operations with non-linear functions, downsampling operations, and classification layer [2,3,17,18].”
BackgroundMethodResultInfluentialA Novel Semisupervised Deep Learning Method for Human Activity Recognition
Zhu, Q., Chen, Z., & Soh, Y. C. (2019) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2018.2889315 · Cited by 97
“Another combination of deep convolutional and LSTM neural networks is also proposed to obtain high-level features on acceleration data [26].”
BackgroundInfluentialIndividual Cattle Identification Using a Deep Learning Based Framework
Qiao, Y., Su, D., Kong, H., Sukkarieh, S., Lomax, S., & Clark, C. (2019) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2019.12.558 · Cited by 122
Li, B., & Sano, A. (2020) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3397318 · Cited by 95
“Similarly, in essence, [66] integrated CNN and long short-term memory (LSTM) networks for human activity recognition based on 500 ms data segmentation from accelerometers and gyroscopes, and [44] also applied CNN-LSTM to recognize emotional levels using 45-minute recordings from smart-phones and…”
MethodKim, K., & Cho, Y. K. (2020) · Automation in Construction · doi:10.1016/j.autcon.2020.103126 · Cited by 100
DeepML: Deep LSTM for Indoor Localization with Smartphone Magnetic and Light Sensors
Wang, X., Yu, Z., & Mao, S. (2018) · International Conference on Communications · doi:10.1109/icc.2018.8422562 · Cited by 96
“The deep LSTM network has been successfully employed for speech recognition [24] and human activity recognition [25].”
MethodZhou, Z., Yu, H., & Shi, H. (2020) · IEEE Access · doi:10.1109/access.2020.2992584 · Cited by 93
“Traditionally, human activity recognition (HAR) [18] tasks have been solved by using engineered features derived from heuristic processes.”
BackgroundEvaluating fusion of RGB-D and inertial sensors for multimodal human action recognition
Imran, J., & Raman, B. (2019) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-019-01239-9 · Cited by 116
“…like accelerometers, gyroscopes and magnetometers have been actively used in the past in areas like action classification (Altun and Barshan 2010; Ordóñez and Roggen 2016), gesture recognition (Lefebvre et al. 2013; Li et al. 2018a), fall detection (Li et al. 2009), sports activities (Ermes et…”
BackgroundMethodLee, D., & Kim, K. (2020) · Renewable Energy · doi:10.1016/j.renene.2020.12.021 · Cited by 93
Pedestrian and Cyclist Detection and Intent Estimation for Autonomous Vehicles: A Survey
Ahmed, S., Huda, M. N., Rajbhandari, S., Saha, C., Elshaw, M., & Kanarachos, S. A. (2019) · Applied Sciences · doi:10.3390/app9112335 · Cited by 112
ProtoHAR: Prototype Guided Personalized Federated Learning for Human Activity Recognition
Cheng, D., Zhang, L., Bu, C., Wang, X., Wu, H., & Song, A. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3275438 · Cited by 92
“During the past decade, several mainstream deep neural networks such as convolutional networks, LSTM, residual networks, and autoencoders [2], [22], [23] have been widely applied to address sensor-based HAR problem.”
BackgroundWearable sensor-based pattern mining for human activity recognition: deep learning approach
Bijalwan, V., Semwal, V. B., & Gupta, V. (2021) · Industrial Robot the international journal of robotics research and application · doi:10.1108/ir-09-2020-0187 · Cited by 102
Dilated causal convolution with multi-head self attention for sensor human activity recognition
Hamad, R. A., Kimura, M., Yang, L., Woo, W. L., & Wei, B. W. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-06007-5 · Cited by 109
Human activity classification based on sound recognition and residual convolutional neural network
Jung, M., & Chi, S. (2020) · Automation in Construction · doi:10.1016/j.autcon.2020.103177 · Cited by 100
Khodairy, M. A., & Abosamra, G. (2021) · IEEE Access · doi:10.1109/access.2020.3048915 · Cited by 97
“Recently, RNNs have been considered one of the most highly utilized models in time-series classification, as they have been achieving state-of-the-art results in several time-series classification tasks such as driver iden-tification [38] and activity recognition [39].”
BackgroundDetection of Parkinson’s disease from handwriting using deep learning: a comparative study
Taleb, C., Likforman-Sulem, L., Mokbel, C., & Khachab, M. Y. (2020) · Evolutionary Intelligence · doi:10.1007/s12065-020-00470-0 · Cited by 91
“Moreover, the CNN-BLSTM model takes advantage of learning the temporal feature activation dynamics [20].”
MethodOnline Fall Detection Using Recurrent Neural Networks on Smart Wearable Devices
Musci, M., De Martini, D., Blago, N., Facchinetti, T., & Piastra, M. (2021) · IEEE Transactions on Emerging Topics in Computing · doi:10.1109/tetc.2020.3027454 · Cited by 128
“n of the classifier. Indeed, they investigate the use of ANN, KNN, Radial Basis Function (RBF), Probabilistic Principal Component Analysis (PPCA) and Linear Discriminant Analysis (LDA). The authors of [14] utilize a DL technique to detect falls. The architecture is composed by two parts: a Convolutional Network and a R…”
MethodDeep learning for the internet of things: Potential benefits and use-cases
Saleem, T. J., & Chishti, M. A. (2020) · Digital Communications and Networks · doi:10.1016/j.dcan.2020.12.002 · Cited by 108
Action Detection and Recognition in Continuous Action Streams by Deep Learning-Based Sensing Fusion
Dawar, N., & Kehtarnavaz, N. (2018) · IEEE Sensors Journal · doi:10.1109/jsen.2018.2872862 · Cited by 99
“In [24], CNN and LSTM layers were combined to achieve action recognition using information from multiple wearable inertial sensors.”
MethodArtificial Intelligence and IoT in Elderly Fall Prevention: A Review
Mohan, D., Al-Hamid, D. Z., Chong, P. H. J., Sudheera, K. L. K., Gutiérrez, J., Chan, H. C. B., & Li, H. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3344605 · Cited by 113
“Camera systems, motion detectors, and IR sensors, as explored in [50], [53], [55], [56], [60], [61], [62], [66], [77], and [78], can be strategically placed to continuously monitor the behavior and gait patterns of the elderly.”
BackgroundWearable Sensor-Based Human Activity Recognition with Hybrid Deep Learning Model
Luwe, Y. J., Lee, C. P., & Lim, K. M. (2022) · Informatics · doi:10.3390/informatics9030056 · Cited by 92
“It can be summarized that the accuracy of deep learning techniques is sorted from low to high as 79.86%, 82.50%, 83.30%, 85.59%, 85.75%, 89.00% and 89.08% of BERT [28], CAE [29], multi-task self-supervision CNN [30], DeepConvLSTM [31], CNN [32], DCNN [33] and CPC [34], in ascending order.”
BackgroundMethodInfluentialDiagnosis and monitoring of Alzheimer's patients using classical and deep learning techniques
Raza, M., Awais, M., Ellahi, W., Aslam, N., Nguyen, H. X., & Le-Minh, H. (2019) · Expert Systems with Applications · doi:10.1016/j.eswa.2019.06.038 · Cited by 88
“Recently, many systems have been developed to objectively classify the ADLs (Awais et al., 2016; Ayachi et al., 2016; Ordóñez & Roggen, 2016).”
BackgroundZhao, J., & Obonyo, E. (2020) · Advanced Engineering Informatics · doi:10.1016/j.aei.2020.101177 · Cited by 98
Sun, J., Fu, Y., Li, S., He, J., Xu, C., & Tan, L. (2018) · Journal of Sensors · doi:10.1155/2018/8580959 · Cited by 88
“Ordóñez and Roggen [18] proposed deep convolutional network with utilization of CNN and LSTM.”
BackgroundMethodInfluentialRecognition and Repetition Counting for Complex Physical Exercises with Deep Learning
Soro, A., Brunner, G., Tanner, S., & Wattenhofer, R. P. (2019) · Sensors · doi:10.3390/s19030714 · Cited by 110
COCOA: Cross Modality Contrastive Learning for Sensor Data
Deldari, S., Xue, H., Saeed, A., Smith, D. V., & Salim, F. D. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3550316 · Cited by 92
“For the sake of completeness and providing a supervised reference, we compare the performance of COCOA to the supervised baseline publicly accepted model, DeepConvLSTM [55] which we call it DeepCL, and a network with the same architecture as COCOA which we call it “sup.” for short.”
MethodWorker 4.0: The Future of Sensored Construction Sites
Calvetti, D., Mêda, P., Gonçalves, M. C., & Sousa, H. J. C. D. (2020) · Buildings · doi:10.3390/buildings10100169 · Cited by 115
Cross-Dataset Activity Recognition via Adaptive Spatial-Temporal Transfer Learning
Qin, X., Chen, Y., Wang, J., & Yu, C. (2019) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3369818 · Cited by 87
“ASTTL is a generic transfer learning method which allows it to be easily extended to deep model by using the deep features extracted by deep neural networks such as CNNs [59] and LSTM [36].”
BackgroundOn the role of features in human activity recognition
Haresamudram, H., Anderson, D. V., & Plötz, T. (2019) · Proceedings of the 23rd International Symposium on Wearable Computers · doi:10.1145/3341163.3347727 · Cited by 90
“We also compare overall classification capabilities of the explored feature representations within the ARC paradigm [6] to a state-of-the-art end-to-end HAR system that does not separate representation learning from classifier training (DeepConvLSTM [34]).”
BackgroundMethodInfluentialC3D-ConvLSTM based cow behaviour classification using video data for precision livestock farming
Qiao, Y., Guo, Y., Yu, K., & He, D. (2022) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106650 · Cited by 97
Arvin, R., Khattak, A. J., & Qi, H. (2020) · Accident Analysis & Prevention · doi:10.1016/j.aap.2020.105949 · Cited by 92
Wu, Z., Guo, Y., Lin, W., Yu, S., & Ji, Y. (2018) · Sensors · doi:10.3390/s18041096 · Cited by 91
“DeepConvLSTM [24] is a kind of DNN, which comprises convolutional, recurrent and softmax layers.”
BackgroundRecognition of human activity using GRU deep learning algorithm
Mohsen, S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15571-y · Cited by 104
“Ordonez and Roggen [40] proposed a ConvLSTM algo-rithm to classify human activities.”
BackgroundMethodInfluentialLearning inverse dynamics models in O(n) time with LSTM networks
Rueckert, E., Nakatenus, M., Tosatto, S., & Peters, J. (2017) · IEEE-RAS International Conference on Humanoid Robots (Humanoids) · doi:10.1109/humanoids.2017.8246965 · Cited by 101
Sanhudo, L., Calvetti, D., Martins, J. P., Ramos, N., Mêda, P., Gonçalves, M. C., & Sousa, H. J. C. D. (2020) · Journal of Building Engineering · doi:10.1016/j.jobe.2020.102001 · Cited by 109
Enhanced Hand-Oriented Activity Recognition Based on Smartwatch Sensor Data Using LSTMs
Mekruksavanich, S., Jitpattanakul, A., Youplao, P., & Yupapin, P. (2020) · Symmetry · doi:10.3390/sym12091570 · Cited by 85
Arefnezhad, S., Samiee, S., Eichberger, A., Frühwirth, M., Kaufmann, C., & Klotz, E. (2020) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.113778 · Cited by 93
Survey on Human Activity Recognition based on Acceleration Data
Slim, S. O., Atia, A., Marwa, M. A., & Mostafa, M.-S. M. (2019) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2019.0100311 · Cited by 97
“Exercise/fitness Walking in treadmill, Running in treadmill, Aerobic dancing, Jumping, Jogging, Playing basketball, Playing football, Rowing [23][24][26][37][42][45][47][48][50] [53] [56][68][70][75][77]”
MethodInfluentialLatent Independent Excitation for Generalizable Sensor-based Cross-Person Activity Recognition
Qian, H., Pan, S. J., & Miao, C. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i13.17416 · Cited by 85
“…transfer learning across domains, many approaches train a deep model on source domains and fine tune with fixed or sequential training data from the target domain (Morales and Roggen 2016b; Buffelli and Vandin 2020; Rokni, Nourollahi, and Ghasemzadeh 2018; Mazankiewicz, B¨ohm, and Berg´es 2020).”
BackgroundMethodInfluentialContrastive Self-supervised Learning for Sensor-based Human Activity Recognition
Khaertdinov, B., Ghaleb, E., & Asteriadis, S. (2021) · 2021 IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb52358.2021.9484410 · Cited by 95
“While plenty of modern deep learning methods have been used for sensor-based HAR in a supervised manner [24, 5, 15, 28, 26, 11, 13], only a few works have been focused on self-supervised learning.”
MethodSemi-supervised convolutional neural networks for human activity recognition
Zeng, M., Yu, T., Wang, X., Nguyen, L. T., Mengshoel, O. J., & Lane, I. (2017) · 2017 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata.2017.8257967 · Cited by 85
“Among supervised methods, deep neural networks (DNNs) have emerged as a method with much potential, in that they are less dependent on clever feature engineering and has strong generalization ability [1] compared to other supervised methods [2], [3].”
BackgroundMethodA Survey on Anomalous Behavior Detection for Elderly Care Using Dense-Sensing Networks
Deep, S., Zheng, X., Karmakar, C. K., Yu, D., Hamey, L., & Jin, J. (2019) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2019.2948204 · Cited by 88
“In addition to the solutions based on traditional machine learning methods, recent studies start using deep learning methods for HAR such as Convolutional Neural Network (CNN) [123], [124] or Recurrent Neural Network (RNN) [125], [126], which represented a better performance than traditional machine learning methods i…”
BackgroundA Survey of Deep Learning Techniques for Cybersecurity in Mobile Networks
Rodríguez, E., Otero, B., Gutiérrez, N., & Canal, R. (2021) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3086296 · Cited by 105
“LSTM has been successfully used in many applications in the areas of speech recognition [92], wearable activity recognition [93] and text categorization [94].”
MethodInterpretable and Accurate Convolutional Neural Networks for Human Activity Recognition
Kim, E. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.2972628 · Cited by 84
“We split the data into training, validation, and test data in the same way as the OPPORTUNITY challenge [13], [18], [30].”
BackgroundMethodResultInfluentialHernandez, V., Dadkhah, D., Babakeshizadeh, V., & Kulić, D. (2020) · Gait & Posture · doi:10.1016/j.gaitpost.2020.10.026 · Cited by 102
Kreuzer, D., Münz, M., & Schlüter, S. (2020) · Machine Learning with Applications · doi:10.1016/j.mlwa.2020.100007 · Cited by 96
Deep Learning Framework for Vehicle and Pedestrian Detection in Rural Roads on an Embedded GPU
Barba-Guaman, L., Eugenio Naranjo, J., & Ortiz, A. (2020) · Electronics · doi:10.3390/electronics9040589 · Cited by 82
“Some deep neural network architectures have been proposed in the past [9], however, we have focused our implementation on the latest architectures that are rapid and well described in [46,47].”
MethodPiotrowski, A., Napiorkowski, J. J., & Piotrowska, A. E. (2019) · Earth-Science Reviews · doi:10.1016/j.earscirev.2019.103076 · Cited by 94
Privacy-Preserving Collaborative Deep Learning with Application to Human Activity Recognition
Lyu, L., He, X., Law, Y. W., & Palaniswami, M. (2017) · Proceedings of the 2017 ACM on Conference on Information and Knowledge Management · doi:10.1145/3132847.3132990 · Cited by 81
“Deep learning techniques have been shown to outperform the existing well-established methods that rely on hand-crafted feature extraction and shallow feature learning architectures, owing to its potential to uncover features that are tied to the dynamics of human motion production, from simple motion encoding in lower…”
Wang, G., Li, Q., Wang, L., Wang, W., Wu, M., & Liu, T. (2018) · Sensors · doi:10.3390/s18061965 · Cited by 99
Cheng, X., Zhang, L., Tang, Y., Liu, Y., Wu, H., & He, J. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3149337 · Cited by 82
“[19] proposed an architecture of DeepConvLSTM, which replaced the fully connected layer of CNN with Long Short Term Memory (LSTM) to capture temporal relationship contained in time series sensor data.”
MethodActivity Recognition for Ambient Assisted Living with Videos, Inertial Units and Ambient Sensors
Ranieri, C. M., MacLeod, S., Dragone, M., Vargas, P. A., & Romero, R. A. F. (2021) · Sensors · doi:10.3390/s21030768 · Cited by 101
“DL architectures, such as Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), autoencoders, Restricted Boltz-mann Machines (RBM), and Recurrent Neural Networks (RNN) have also been successfully applied to this modality [33].”
MethodAbidine, M. B., Fergani, L., Fergani, B., & Oussalah, M. (2016) · Pattern Analysis and Applications · doi:10.1007/s10044-016-0570-y · Cited by 90
“Ordonez and Roggen [28] suggested a convolutional and LSTM recurrent units to recognize multimodal wearable activities.”
BackgroundBig Healthcare Data Analytics: Challenges and Applications
Lee, C., Luo, Z., Ngiam, K. Y., Zhang, M., Zheng, K., Chen, G., Ooi, B. C., & Yip, W. L. J. (2017) · Scalable computing and communications · doi:10.1007/978-3-319-58280-1_2 · Cited by 118
“In [83], a deep learning model is designed using convolutional neural networks and LSTM recurrent neural networks, which captures spatiotemporal patterns of signals from wearable accelerometers and gyroscopes.”
MethodHybrid Convolutional Neural Network-Multilayer Perceptron Model for Solar Radiation Prediction
Ghimire, S., Nguyen‐Huy, T., Prasad, R., Deo, R. C., Casillas-Pérez, D., Salcedo‐Sanz, S., & Bhandari, B. (2022) · Cognitive Computation · doi:10.1007/s12559-022-10070-y · Cited by 95
“D domain, a filtering kernel plays as a feature extractor with one-dimensional convolution described in [60]: where a ( l + 1 ) j refers to the feature map j in layer l + 1 , 𝜎 refers to a nonlinear function, F l indicates the number of feature (8) Fig.”
BackgroundKumaran Santhosh, K., Dogra, D. P., Roy, P. P., & Mitra, A. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3108504 · Cited by 80
Human activity recognition using marine predators algorithm with deep learning
Helmi, A. M., Al‐qaness, M. A. A., Dahou, A., & Elaziz, M. A. (2023) · Future Generation Computer Systems · doi:10.1016/j.future.2023.01.006 · Cited by 85
Hong, S., Zhou, Y., Wu, M., Shang, J., Wang, Q., Li, H., & Xie, J. (2019) · Physiological Measurement · doi:10.1088/1361-6579/ab15a2 · Cited by 79
Automated Stoichiometry Analysis of Single-Molecule Fluorescence Imaging Traces via Deep Learning
Xu, J., Qin, G., Luo, F., Wang, L., Zhao, R., Li, N., Yuan, J., & Fang, X. (2019) · Journal of the American Chemical Society · doi:10.1021/jacs.9b00688 · Cited by 81
A bibliometric analysis on deep learning during 2007–2019
Li, Y., Xu, Z., Wang, X., & Wang, X. (2020) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-020-01152-0 · Cited by 105
Medina-Quero, J., Zhang, S., Nugent, C., & Espinilla, M. (2018) · Expert Systems with Applications · doi:10.1016/j.eswa.2018.07.068 · Cited by 85
Shuvo, M. M. H., Ahmed, N., Nouduri, K., & Palaniappan, K. (2020) · IEEE Applied Imagery Pattern Recognition Workshop · doi:10.1109/aipr50011.2020.9425332 · Cited by 98
“SVM [3] 96.37 SVM+HMM [17] 97.60 LSTM-CNN [18] 95.78 Bidirectional-LSTM [19] 92.67 CNN-LSTM [20] 93.40 1D CNN [5] 94.79 DeepCNN [2] 97.59 TFFT + CNN [5] 95.75 DeepConvLSTM [21] 95.80 Stacked Autoencoder [22] 92.16 CHMM [23] 93.18 Proposed hybrid method 97.66”
InfluentialTinyHAR: A Lightweight Deep Learning Model Designed for Human Activity Recognition
Zhou, Y., Zhao, H., Huang, Y., Riedel, T., Hefenbrock, M., & Beigl, M. (2022) · Proceedings of the 2022 ACM International Symposium on Wearable Computers · doi:10.1145/3544794.3558467 · Cited by 78
“DeepConvLSTM [21], a popular benchmark HAR model, first used a CNN subnet to extract local features from different sensor modalities and feed the extracted features into a long-short-term-memory (LSTM).”
BackgroundMethodAccelerometer-Based Human Activity Recognition for Patient Monitoring Using a Deep Neural Network
Fridriksdottir, E., & Bonomi, A. (2020) · Sensors · doi:10.3390/s20226424 · Cited by 88
Deep learning-based classification of work-related physical load levels in construction
Yang, K., Ahn, C. R., & Kim, H. (2020) · Advanced Engineering Informatics · doi:10.1016/j.aei.2020.101104 · Cited by 87
Sensor Data Augmentation by Resampling in Contrastive Learning for Human Activity Recognition
Wang, J., Zhu, T., Gan, J., Chen, L. L., Ning, H., & Wan, Y. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3214198 · Cited by 76
Baldominos, A., Cervantes, A., Sáez, Y., & Isasi, P. (2019) · Sensors · doi:10.3390/s19030521 · Cited by 86
RNN-Based Path Prediction of Obstacle Vehicles With Deep Ensemble
Min, K., Kim, D., Park, J., & Huh, K. (2019) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2019.2933232 · Cited by 83
“[23] performed human activity recognition using a motion sensor by 1D convolution and RNN.”
MethodDeep Transfer Learning for Time Series Data Based on Sensor Modality Classification
Li, F., Shirahama, K., Nisar, M. A., Huang, X., & Grzegorzek, M. (2020) · Sensors · doi:10.3390/s20154271 · Cited by 74
Tsinganos, P., & Skodras, A. N. (2018) · Sensors · doi:10.3390/s18020592 · Cited by 85
Park, S.-Y., & Ahn, J. (2020) · Acta Astronautica · doi:10.1016/j.actaastro.2020.08.019 · Cited by 80
Impact of Physical Activity on Sleep:A Deep Learning Based Exploration
Sathyanarayana, A., Joty, S., Fernández-Luque, L., Ofli, F., Srivastava, J., Elmagarmid, A. K., Taheri, S., & Arora, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.07034 · Cited by 74
“speech recognition, Ordonez and Roggen recently proposed an architecture comprising both convolutional and LSTM recurrent neural networks for activity recognition from multimodal wearable sensor data [36]. The convolutional layers in this architecture acted as feature extractors and provided abstract representations o…”
BackgroundImproving the prediction of continuous integration build failures using deep learning
Saidani, I., Ouni, A., & Mkaouer, M. W. (2022) · Automated Software Engineering · doi:10.1007/s10515-021-00319-5 · Cited by 94
Iss2Image: A Novel Signal-Encoding Technique for CNN-Based Human Activity Recognition
Hur, T., Bang, J., Huynh-The, T., Lee, J., Kim, J.-I., & Lee, S. (2018) · Sensors · doi:10.3390/s18113910 · Cited by 74
Deep Temporal Multimodal Fusion for Medical Procedure Monitoring Using Wearable Sensors
Bernal, E. A., Yang, X., Li, Q., Kumar, J., Madhvanath, S., Ramesh, P., & Bala, R. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2017.2726187 · Cited by 74
“A recent example of a deep multimodal data fusion framework for action recognition that does not rely on hand-engineered features is [68]; although the proposed method includes temporal modeling of the data, the fusion is performed with atemporal feedforward networks, more specifically CNNs, so the temporal”
MethodImproving Deep Learning for HAR with Shallow LSTMs
Bock, M., Hölzemann, A., Moeller, M., & Van Laerhoven, K. (2021) · 2021 International Symposium on Wearable Computers · doi:10.1145/3460421.3480419 · Cited by 74
“Our paper’s contributions are threefold: (1) We show that an altered DeepConvLSTM architecture with a one-layered LSTM overall outperforms architectures employing a two-layered LSTM, by validating our claim using the Opportunity dataset [29] as seen in [26], as well as 4 other popular HAR datasets [28, 32, 33, 36].”
BackgroundMethodInfluentialCross-Subject and Cross-Modal Transfer for Generalized Abnormal Gait Pattern Recognition
Gu, X., Guo, Y., Deligianni, F., Lo, B. P. L., & Yang, G. (2020) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.3009448 · Cited by 83
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.