Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Open accessThe DOI matches its Crossref record; 49 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., and Li, F.F. (2015). Visualizing and understanding recurrent networks. Cornell Univ. Lab., 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., and Dambre, J. (2015). Beyond Temporal Pooling: Recurrence and Temporal Convolutions for Gesture Recognition in Video. arXiv Preprint, 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
- LeCun, Y., and Bengio, Y. (1998). The Handbook of Brain Theory and Neural Networks, MIT Press.
Cited by
Items 1,201–1,300 of 2,806
Recognizing Grabbing Actions from Inertial and Video Sensor Data in a Warehouse Scenario
Diete, A., Sztyler, T., Weiland, L. R., & Stuckenschmidt, H. (2017) · Procedia Computer Science · doi:10.1016/j.procs.2017.06.071 · Cited by 15
Patidar, P., Goel, M., & Agarwal, Y. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3610907 · Cited by 20
“Building upon these approaches, multi-modal sensors and sensor fusion has also been proposed for more accurate HAR [4, 68, 73, 75, 78].”
MethodLog-Viterbi algorithm applied on second-order hidden Markov model for human activity recognition
Yang, S., Thapa, K., Kabir, M. H., & Hee-Chan, L. (2018) · International Journal of Distributed Sensor Networks · doi:10.1177/1550147718772541 · Cited by 13
An intelligent collaboration framework of IoT applications based on event logic graph
Zhu, M., Han, Y., Liu, Z., et al. (2022) · Future Generation Computer Systems · doi:10.1016/j.future.2022.06.017 · Cited by 12
Bangaru, S. S., Wang, C., Aghazadeh, F., Muley, S., & Willoughby, S. (2025) · Sensors · doi:10.3390/s25103204 · Cited by 12
“Recurrent Neural Networks (RNNs) such as Long Short-Term Memory (LSTM) have been effective for sequence data and regression [50,51].”
Activity recognition through interactive machine learning in a dynamic sensor setting
Tegen, A., Davidsson, P., & Persson, J. (2020) · Personal and Ubiquitous Computing · doi:10.1007/s00779-020-01414-2 · Cited by 13
“introduce a framework for activity recognition based on convolutional and LSTM recurrent units and test it on two benchmark datasets [20].”
Deep Multimodal Representation Learning from Temporal DataPreprint
Yang, X., Ramesh, P. S., Chitta, R., et al. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.03152 · Cited by 12
“…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.”
MethodReal-Time Human Activity Recognition on Embedded Equipment: A Comparative Study
Najeh, H., Lohr, C., & Leduc, B. (2024) · Applied Sciences · doi:10.3390/app14062377 · Cited by 18
Alzeyadi, M. K. K., Der, C. S., & Chen, C. P. (2025) · Engineering Research Express · doi:10.1088/2631-8695/ada72d · Cited by 12
Energy-efficient neural networks with near-threshold processors and hardware accelerators
Nunez-Yanez, J. L., & Howard, N. (2021) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2021.102062 · Cited by 13
Industrial robot programming by demonstration using stereoscopic vision and inertial sensing
de Souza, J. P. C., Amorim, A. M., Rocha, L. F., et al. (2021) · Industrial Robot the international journal of robotics research and application · doi:10.1108/ir-02-2021-0043 · Cited by 12
Ohno, K., & Kumagai, A. (2021) · IEEE International Conference on Big Knowledge (ICBK) · doi:10.1109/ickg52313.2021.00033 · Cited by 18
“Recurrent Neural Networks (RNNs) are deep learning models for representing sequential data, which have extensive applications including speech recognition [1], [2], natural language processing [3], video analysis [4], and action recognition [5].”
BackgroundPosture Recognition Technology Based on Kinect
Li, Y., Zhijie, C. H. U., & Xin, Y. (2020) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2019edp7221 · Cited by 12
“Deep learning [41] method was di ffi cult to distinguish the overlapping of motions, and the recognition accuracy was often a ff ected for reciprocating motions.”
BackgroundMethodGazzari, M., Mattmann, A., Maaß, M., & Hollick, M. (2021) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3494986 · Cited by 12
“This particular network is inspired by works in the related field of human activity recognition (HAR) [2, 25], dealing with a related task of classifying multivariate sensor data.”
BackgroundDeep Learning-Based Recognition of Unsafe Acts in Manufacturing Industry
Vukićević, A. M., Petrović, M., Knežević, N., & Jovanovic, K. M. (2023) · IEEE Access · doi:10.1109/access.2023.3318114 · Cited by 14
Uzunhisarcıklı, E., Kavuncuoğlu, E., & Özdemir, A. T. (2022) · Computational Intelligence · doi:10.1111/coin.12517 · Cited by 13
“36,41,42 The difference between the repeating module in the LSTM structure is that four layers are connected uniquely instead of a single neural network layer.”
BackgroundAn intelligent well-being monitoring system for residents in extra care homes
Jing, Y., Eastwood, M., Tan, B., Konios, A., Hamid, A., & Collinson, M. (2017) · Proceedings of the 1st International Conference on Internet of Things and Machine Learning · doi:10.1145/3109761.3109769 · Cited by 12
Multi-Sensor Time-Series Classification for Activity Tracking Under Variable Length
Norgaard, S., Saeedi, R., & Gebremedhin, A. H. (2019) · IEEE Sensors Journal · doi:10.1109/jsen.2019.2953938 · Cited by 13
Mani, N., Haridoss, P., & George, B. (2023) · IEEE Open Journal of Instrumentation and Measurement · doi:10.1109/ojim.2023.3301861 · Cited by 14
“The selection of discriminatory analysis and the choice of machine learning models for evaluation is based on the models that were successful and widely used for HAR problems [1], [5], [8], [16], [19], [20], [21], [22], [23], [24], [25], [26], [27].”
MethodMonitoring hand hygiene with commercial gas sensors: A pattern recognition approach
Braun, S., Kobald, A., Oprea, A., Boehme, I., Bonanati, P., Weimar, U., & Bârsan, N. (2021) · Sensors and Actuators B Chemical · doi:10.1016/j.snb.2021.131027 · Cited by 12
Deep Convolutional Bidirectional LSTM for Complex Activity Recognition with Missing Data
Saha, S. S., Sandha, S. S., & Srivastava, M. (2021) · Smart Innovation, Systems and Technologies · doi:10.1007/978-981-15-8269-1_4 · Cited by 12
Applying Deep Learning Techniques to Estimate Patterns of Musical Gesture
Dalmazzo, D., Waddell, G., & Ramírez, R. (2021) · Frontiers in Psychology · doi:10.3389/fpsyg.2020.575971 · Cited by 13
“Ordóñez and Roggen (2016) have implemented a 3DCNN-LSTM deep neural network to recognize seven gestures (five for finger tapping, one palm tapping, and one pointing) as a model for interactive music applications utilizing the LeapMotion device.”
BackgroundMethodMachine learning models for wearable-based human activity recognition: A comparative study
Ciortuz, G., Pour, H. H., Irshad, M. T., Nisar, M. A., Huang, X., & Fudickar, S. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130911 · Cited by 18
Jansi, R., & Amutha, R. (2021) · Swarm and Evolutionary Computation · doi:10.1016/j.swevo.2021.100873 · Cited by 12
Classifying Goliath Grouper (Epinephelus itajara) Behaviors from a Novel, Multi-Sensor Tag
Brewster, L. R., Ibrahim, A. K., DeGroot, B. C., Ostendorf, T. J., Zhuang, H., Chérubin, L. M., & Ajemian, M. J. (2021) · Sensors · doi:10.3390/s21196392 · Cited by 12
Valarezo, A., Rivera, L., Park, H., Park, N., & Kim, T.-S. (2020) · Computer Science and Information Systems · doi:10.2298/csis190920005v · Cited by 12
“, inference on Android device), while most of the previous HAR systems are implemented only for PC [29] [28] [27].”
BackgroundHatkeposhti, R. K., Yadollahzadeh‐Tabari, M., & GolsorkhTabariAmiri, M. (2022) · International Journal of Engineering · doi:10.5829/ije.2022.35.10a.13 · Cited by 15
Tian, Y., Hettiarachchi, D., Yu, H. S., & Kamijo, S. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3404777 · Cited by 15
“Ordóñez and Roggen [28] proposed a method called DeepConvLSTM (DCL) composed of CNN layers and LSTM layer to learn the time distribution, it feeds the raw time-domain data from multiple sensors to the CNN layer, and the LSTM layer further processes the output of the convolutional layer.”
MethodHurley, N. C., Spatz, E. S., Krumholz, H. M., Jafari, R., & Mortazavi, B. J. (2020) · ACM Transactions on Computing for Healthcare · doi:10.1145/3417958 · Cited by 14
“Multiple-sensor fusion approaches have improved a number of modeling tasks, including estimating of blood pressure from ECG and PPG signals [123] and HAR through the use of inertial measurement units that integrate accelerometers, gyroscopes, and magnetometers with use of multiple sensors across the body [144].”
BackgroundEspialCog: General, Efficient and Robust Mobile User Implicit Authentication in Noisy Environment
Zhu, T., Weng, Z., Song, Q., Yuan, C., Liu, Q., Yan, C., Lv, M., & Chen, T. (2020) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2020.3012491 · Cited by 11
“Few works used LSTM for motion sensor-based user authentication, others such as [28, 29] utilized LSTM for wearable activity recognition, but the noisy labels in the complex environment and a large amount of real-world data were not considered by them.”
Activity Semantics-Based Indoor Localization Using Smartphones
Zhou, B., Wu, P., Zhang, X., Zhang, D., & Li, Q. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3357718 · Cited by 15
“Ordóñez et al. [24] proposed an activity detection framework based on a convolutional neural network (CNN) and a long short-term memory (LSTM) network, and the accuracy they achieved on the OPPORTUNITY dataset was as high as 91.5%.”
MethodInfluentialSensor Positioning and Data Acquisition for Activity Recognition using Deep Learning
Chung, S., Lim, J., Noh, K. J., Kim, G. G., & Jeong, H. T. (2018) · International Conference on Information and Communication Technology Convergence (ICTC) · doi:10.1109/ictc.2018.8539473 · Cited by 13
Continual Egocentric Activity Recognition With Foreseeable-Generalized Visual–IMU Representations
He, C., Cheng, S., Qiu, Z., Xu, L., Meng, F., Wu, Q., & Li, H. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3371975 · Cited by 11
“For these captured IMU sensing sequences, the processing is generally performed using LSTM-based models [28], [29].”
MethodExploratory Analysis of Smartphone Sensor Data for Human Activity Recognition
Islam, S. M. M., & Talukder, K. H. (2023) · IEEE Access · doi:10.1109/access.2023.3314651 · Cited by 12
Lian, C., Zhao, Y., Sun, T., Shao, J., Liu, Y., Fu, C., Lyu, X., & Zhan, Z. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113076 · Cited by 13
Compare of Machine Learning and Deep Learning Approaches for Human Activity Recognition
Moradi, B., Aghapour, M., & Shirbandi, A. (2022) · Iranian Conference on Electrical Engineering · doi:10.1109/icee55646.2022.9827335 · Cited by 16
(2017) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-72038-8 · Cited by 11
Yu, J., Zhang, L., Cheng, D., Bu, C., Wu, H., & Song, A. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3412736 · Cited by 15
“Ordóñez and Roggen [20] developed an automatic feature extraction system named DeepConvLSTM.”
MethodKumar, P., & Chauhan, S. (2021) · CCF Transactions on Pervasive Computing and Interaction · doi:10.1007/s42486-021-00063-5 · Cited by 13
RetractedAdaptive Multi-Scale Wavelet Neural Network for Time Series Classification
Ouyang, K., Hou, Y., Zhou, S., & Zhang, Y. (2021) · Information · doi:10.3390/info12060252 · Cited by 18
“Time series data from the accelerometers, gyroscopes, or magnetic field sensors is used to recognize the human activity recognition [2].”
MethodEmMixformer: Mix Transformer for Eye Movement Recognition
Qin, H., Zhu, H., Jin, X., Song, Q., El‐Yacoubi, M. A., & Gao, X. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3551452 · Cited by 28
“The LSTM is widely used for sequence processing [54], [55].”
Deriving Effective Human Activity Recognition Systems through Objective Task Complexity Assessment
Hiremath, S. K., & Plötz, T. (2020) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3432227 · Cited by 11
DanHAR: Dual Attention Network For Multimodal Human Activity Recognition Using Wearable SensorsPreprint
Gao, W., Zhang, L., Teng, Q., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.14435 · Cited by 11
Recurrent Transformation of Prior Knowledge Based Model for Human Motion Recognition
Xu, C., He, J., Zhang, X., Cai, H., Duan, S., Tseng, P., & Li, C. (2018) · Computational Intelligence and Neuroscience · doi:10.1155/2018/4160652 · Cited by 13
“For example, Ordóñez and Roggen [19] proposed a generic deep framework (DeepConvLSTM) for activity recognition based on convolutional and LSTM recurrent units.”
BackgroundDeep Recurrent Neural Networks for Edge Monitoring of Personal Risk and Warning Situations
Torti, E., Musci, M., Guareschi, F., Leporati, F., & Piastra, M. (2019) · Scientific Programming · doi:10.1155/2019/9135196 · Cited by 11
“Among these methods for fall detection, deep learning techniques recently showed to be a very promising approach [2, 3].”
MethodKing, T., Zhou, Y., Röddiger, T., & Beigl, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-90764-z · Cited by 24
“This two-step process is a common approach in the domain of time series classification 8,19,20,41,42 .”
Human Brain Tissue Segmentation in fMRI using Deep Long-Term Recurrent Convolutional Network
Ang, S. P., Phung, S. L., Schira, M. M., Bouzerdoum, A., & Duong, S. T. M. (2018) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2018.8615850 · Cited by 12
“CNN is the state-of-theart architecture in image recognition [5], and LSTM is widely used for sequence-related problems such as speech recognition [6], human activity recognition [7], and time series classification [8].”
MethodWang, Z., French, N., James, T., Schillaci, C., Chan, F. K. S., Feng, M., & Lipani, A. (2023) · Journal of Sustainable Agriculture and Environment · doi:10.1002/sae2.12041 · Cited by 16
Combining Public Human Activity Recognition Datasets to Mitigate Labeled Data Scarcity
Presotto, R., Ek, S., Civitarese, G., Portet, F., Lalanda, P., & Bettini, C. (2023) · IEEE International Conference on Smart Computing · doi:10.1109/smartcomp58114.2023.00022 · Cited by 24
Unmanned ground vehicle (UGV) based automated construction progress measurement of road using LSTM
Patel, T., Guo, B. H. W., van der Walt, J. D., & Zou, Y. (2024) · Engineering Construction & Architectural Management · doi:10.1108/ecam-01-2024-0020 · Cited by 15
Zahid, H., Syed, S. A., Rashid, M., Hussain, S., Umer, A., Waheed, A., Nasim, S., Zareei, M., & Mansoor, N. (2023) · Mobile Information Systems · doi:10.1155/2023/1060135 · Cited by 13
“In sign recognition, dynamic mundane dependency learning yielded signifcant results [38].”
BackgroundGuo, L., Yang, B., & Ye, J. (2020) · IEEE Transportation Electrification Conference and Expo · doi:10.1109/itec48692.2020.9161645 · Cited by 11
A Review of Physical Human Activity Recognition Chain Using Sensors
AlShorman, O., Alshorman, B., & Masadeh, M. (2020) · Indonesian Journal of Electrical Engineering and Informatics (IJEEI) · doi:10.11591/ijeei.v8i3.2312 · Cited by 20
Real-time recognition of team behaviors by multisensory graph-embedded robot learning
Reily, B., Gao, P., Han, F., Wang, H., & Zhang, H. (2021) · The International Journal of Robotics Research · doi:10.1177/02783649211043155 · Cited by 13
“Specifically, LSTM and RNNs have also been used to identify activities from large sets of noisy data (Ordo´n˜ez and Roggen (2016)).”
MethodRefat, M. A. R., Hossain, M. P., Islam, M. R., Rahman, A., Farid, F. A., Karim, H. B. A., & Miah, A. S. M. (2025) · Discover Applied Sciences · doi:10.1007/s42452-025-07896-0 · Cited by 13
“To achieve a comprehensive comparison, we compared the TECA-HAR model with the following baseline models: CNN [35], vanilla LSTM [31], CNN-LSTM [31], Bidirectional LSTM [32], Bidirectional GRU [32], and Deep Convolutional LSTM [61], T-DCNN (Ours), respectively.”
Unleashing the Power of Shared Label Structures for Human Activity Recognition
Zhang, X., Chowdhury, R. R., Zhang, J., Hong, D., Gupta, R. K., & Shang, J. (2023) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3583780.3615101 · Cited by 26
Polo-Rodríguez, A., Chiachio, J. M. V., Paggetti, C., & Medina-Quero, J. (2021) · Applied Sciences · doi:10.3390/app11156978 · Cited by 13
Lane Change Intention Prediction of CNN-LSTM Based on Multi-head Attention
Kai, G., Li, X., Hu, L., Chen, B., & Du, R. (2022) · Journal of Mechanical Engineering · doi:10.3901/jme.2022.22.369 · Cited by 12
Zhang, K., Liu, S., An, S., Zhou, Y., & Pu, X. (2025) · Nano Energy · doi:10.1016/j.nanoen.2025.111337 · Cited by 13
CNN Real-Time Detection of Vandalism Using a Hybrid -LSTM Deep Learning Neural Networks
Nyajowi, T., Oyie, N. O., & Ahuna, M. (2021) · IEEE AFRICON · doi:10.1109/africon51333.2021.9570902 · Cited by 12
Hu, L., Zhao, K., Ling, B. W., & Lin, Y. (2022) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2022.104255 · Cited by 12
Stacked Convolutional LSTM Models for Prognosis of Bearing Performance Degradation
Wang, D., & Wang, Y. (2019) · Prognostics and System Health Management Conference · doi:10.1109/phm-qingdao46334.2019.8943063 · Cited by 11
“[11] Ordóñez F, Roggen D, "Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition," Sensors 16.”
BackgroundAre Machine Learning Methods the Future for Smoking Cessation Apps?
Abo-Tabik, M., Benn, Y., & Costen, N. P. (2021) · Sensors · doi:10.3390/s21134254 · Cited by 16
Human-Aware Collaborative Robots in the Wild: Coping with Uncertainty in Activity Recognition
Yalçinkaya, B., Couceiro, M. S., Soares, S., & Valente, A. L. G. (2023) · Sensors · doi:10.3390/s23073388 · Cited by 12
Smartphone IMU Sensors for Human Identification through Hip Joint Angle Analysis
Andersson, R., Bermejo-García, J., Agujetas, R., Cronhjort, M. B., & Chilo, J. (2024) · Sensors · doi:10.3390/s24154769 · Cited by 15
“…neural network (RNN), naive Bayes (NB), linear discriminant analysis (LDA), and hybrid convolutional neural network (HCNN), were utilized in previous studies for gait recognition and classification to extract a comprehensive understanding of biometric data based on human movements [9,32,33].”
MethodGuo, X., Kim, Y., Ning, X., & Min, S. D. (2025) · Sensors · doi:10.3390/s25020301 · Cited by 13
“For comparison purposes, we replicate the widely used CNNs-LSTM framework [13] on three benchmark datasets for HAR.”
InfluentialFeature-Level Cross-Attentional PPG and Motion Signal Fusion for Heart Rate Estimation
Kasnesis, P., Toumanidis, L., Burrello, A., Chatzigeorgiou, C., & Patrikakis, C. Z. (2023) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac57700.2023.00267 · Cited by 11
Santhosh, J., Dzsotjan, D., & Ishimaru, S. (2023) · IEEE Access · doi:10.1109/access.2023.3311268 · Cited by 14
“Meanwhile, deep learning models have the benefit of extracting high-level features from raw sensory input and generating better outcomes [25], [26], [27], [28], [29].”
BackgroundSmart-watch-based construction worker activity recognition with hand-held power tools
Altheimer, J., & Schneider, J. (2024) · Automation in Construction · doi:10.1016/j.autcon.2024.105684 · Cited by 20
Dimitrievski, A., Zdravevski, E., Lameski, P., Villasana, M. V., Pires, I. M., García, N. M., Flórez‐Revuelta, F., & Trajkovik, V. (2021) · Sensors · doi:10.3390/s21093030 · Cited by 9
“[39], or by using deep learning approaches, such as that of Ordóñez and Roggen [40].”
MethodStatistical Database of Human Motion Recognition Using Wearable IoT—A Review
Asl, E. F., Ebadollahi, S., Vahidnia, R., & Jalali, A. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3282171 · Cited by 18
“for wearable activity recognition [67].”
BackgroundMethodInfluentialNovel Human Activity Recognition by graph engineered ensemble deep learning model
Ghalan, M., & Aggarwal, R. K. (2024) · IFAC Journal of Systems and Control · doi:10.1016/j.ifacsc.2024.100253 · Cited by 12
Rivera, P., Añazco, E. V., & Kim, T. (2020) · International Journal of Pattern Recognition and Artificial Intelligence · doi:10.1142/s0218001421520066 · Cited by 10
Reminder Care System: An Activity-Aware Cross-Device Recommendation System
Altulyan, M. S., Huang, C., Yao, L., Wang, X., Kanhere, S. S., & Cao, Y. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-35231-8_15 · Cited by 13
“As here the input is time series signals from multiple IMUs, DeepConvLSTM [17] classifiers have been proved powerful in handling such data.”
BackgroundActivity Recognition Using Different Sensor Modalities and Deep Learning
Ascioglu, G., & Şenol, Y. (2023) · Applied Sciences · doi:10.3390/app131910931 · Cited by 12
Seo, S., & Chung, M. (2022) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2022.0130210 · Cited by 10
“CNN, which mainly uses two-dimensional data, can be applied to data feature extraction and data prediction analysis by utilizing one-dimensional time series data [19-21].”
MethodUse of Technologies for the Acquisition and Processing Strategies for Motion Data Analysis
Hurtado-Perez, A. E., Toledano‐Ayala, M., Cruz-Albarrán, I. A., Lopez-Zúñiga, A., Moreno-Perez, J. A., Álvarez-López, A., Rodríguez‐Reséndiz, J., & Perez-Ramirez, C. A. (2025) · Biomimetics · doi:10.3390/biomimetics10050339 · Cited by 16
“…LSTM incorporates internal memory cells and gating mechanisms (input, forget, and output gates) for allowing the network to retain or discard information over time, making them particularly suitable for time-series data such as those generated in motion capture or wearable sensor systems [120].”
Shao, E., Mei, Q., Ye, J., Ugbolue, U. C., Chen, C., & Gu, Y. (2022) · Bioengineering · doi:10.3390/bioengineering9090411 · Cited by 10
Environment Simulation for the Promotion of the Open Data Initiative
Synnott, J., Nugent, C., Zhang, S., et al. (2016) · IEEE International Conference on Smart Computing · doi:10.1109/smartcomp.2016.7501690 · Cited by 10
Yang, L., Li, X., Ma, Z., Li, L., Xiong, N. N., & Ma, J. (2023) · ACM Transactions on Internet Technology · doi:10.1145/3594538 · Cited by 12
“In summary, extensive experiments are conducted to demonstrate the performance of our proposed IRGA system on three datasets (ZJU-GaitAcc [47], MobiAct [40], and our own dataset), and the comparisons with state-of-the-art gait authentication methods (SPs[47], [29], [38], [35], and DeepConvLSTM[33]) are established to…”
BackgroundMethodInfluentialNeogi, S. (2020) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/143222 · Cited by 10
Quality prediction of tractor rotary tillage based on BiConvLSTM with self-attention
Li, R.-C., Wen, C., Li, S., Li, R., Pu, H.-Y., Jiang, Y., & Song, Z. (2023) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2023.107643 · Cited by 12
(2021) · doi:10.1007/s43441-021-00355-z · Cited by 11
“LeNet [54], a CNN-based handwritten number recognition system—was developed and succeeded by a CNN called AlexNet [30], which is a key trigger for renewed interest in neural networks.”
MethodPreliminary Investigation of Visualizing Human Activity Recognition Neural Network
Yoshimura, N., Maekawa, T., & Hara, T. (2019) · International Conference on Mobile Computing and Ubiquitous Network (ICMU) · doi:10.23919/icmu48249.2019.9006643 · Cited by 10
“Various models of wearable activity recognition have been proposed [2].”
BackgroundCNN-based Speed Detection Algorithm for Walking and Running using Wrist-worn Wearable Sensors
Seethi, V. D. R., & Bharti, P. (2020) · IEEE International Conference on Smart Computing · doi:10.1109/smartcomp50058.2020.00064 · Cited by 13
Huynh‐The, T., Hua, C.-H., & Kim, D. (2019) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2019.8857366 · Cited by 14
“, an advancement of RNNs) [14]– [16] have been considered for automatically learning discriminative features from raw sensory data.”
BackgroundBenchmarking Annotation Procedures for Multi-channel Time Series HAR Dataset
Avsar, H., Altermann, E., Reining, C., Rueda, F. M., Fink, G. A., & ten Hompel, M. (2021) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops51409.2021.9431062 · Cited by 16
“Statistical pattern recognition and deep neural networks have been used for solving multi-channel time series HAR [1], [6], [8], [15]–[17].”
BackgroundMethodInfluentialLi, H., Derrode, S., & Wojciech Pieczynski, E. (2019) · Sensors · doi:10.3390/s19194242 · Cited by 12
“In addition, for the OPPORTUNITY dataset, [18] used CNN to obtain an accuracy of 77.99% by using the body-worn sensors only, while in [33], CNN obtained an accuracy of 93.75% on six activities: walking, stair ascent, stair descent, sitting, standing, and laying.”
InfluentialCheng, S., & Liu, Y. (2023) · Sensors · doi:10.3390/s23073585 · Cited by 11
Enokibori, Y. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3699733 · Cited by 17
“Only a few can be verified, including DeepConvLSTM [26], mobileHART [9], and iSPLInception [30].”
Residual Learning and LSTM Networks for Wearable Human Activity Recognition Problem
Yu, S. (2018) · Chinese Control Conference (CCC) · doi:10.23919/chicc.2018.8482968 · Cited by 11
“Some simple activity-aware systems are now commercial in the form of fitness trackers or fall detection devices [5].”
BackgroundDeepContext: Parameterized Compatibility-Based Attention CNN for Human Context Recognition
Alajaji, A. S., Gerych, W., Chandrasekaran, K., Buquicchio, L., Agu, E., & Rundensteiner, E. A. (2020) · IEEE International Conference on Semantic Computing (ICSC) · doi:10.1109/icsc.2020.00015 · Cited by 12
Accurate gait recognition with inertial sensors using a new FCN-BiLSTM architecture
Rifaat, N., Ghosh, U. K., & Sayeed, A. (2022) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2022.108428 · Cited by 19
Masked Video and Body-Worn IMU Autoencoder for Egocentric Action Recognition
Zhang, M., Huang, Y., Liu, R., & Sato, Y. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72649-1_18 · Cited by 31
Personalised Meta-Learning for Human Activity Recognition with Few-Data
Wijekoon, A., & Wiratunga, N. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-63799-6_6 · Cited by 11
Rosario, M. B. D., Lovell, N. H., & Redmond, S. J. (2019) · Sensors · doi:10.3390/s19132845 · Cited by 11
“With this in mind, the state-of-the-art deep learning methods (recall Section 1.3.3) proposed by Ordoñez et al. [45] and Li et al. [78] are excluded because they utilize measurements from multiple IMUs that are placed at different anatomical locations on the body, whilst the works of Ravi et al.…”
BackgroundMethodInfluentialHoang, N. N., Lee, G.-S., Kim, S.-H., et al. (2018) · International Conference on Machine Learning and Machine Intelligence (MLMI) · doi:10.1145/3278312.3278314 · Cited by 10
Hoelzemann, A., & Laerhoven, K. V. (2024) · Frontiers in Computer Science · doi:10.3389/fcomp.2024.1379788 · Cited by 16
“The deep learning analyses are performed using the DeepConvLSTM architecture [11] which is based on a Keras implementation of [16].”
MethodWearable Sensors Based Automatic Box and Block Test System
Zhang, Y., Chen, Y., Yu, H., et al. (2019) · doi:10.1109/smartworld-uic-atc-scalcom-iop-sci.2019.00189 · Cited by 10
“The vectorial representation of the update of an LSTM layer is as follows [30]: 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 value.”
BackgroundWang, J., Cui, W., Zhu, T., Ning, H., & Liu, Z. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3390755 · Cited by 13
“Supervised deep learning methods have been proven to significantly improve the accuracy of activity recognition compared to traditional methods [10], [12], [29], [30], [31].”
Method
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.