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 1,701–1,800 of 2,807
Exploiting Typicality for Selecting Informative and Anomalous Samples in Videos
Bappy, J. H., Paul, S., Tuncel, E., & Roy–Chowdhury, A. K. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2910634 · Cited by 6
“Recently, various deep learning based models have been presented in [18]–[22] for activity classification.”
BackgroundSheng, T., & Huber, M. (2025) · Sensors · doi:10.3390/s25134032 · Cited by 5
“Then, the feature extraction can be defined as Equation (8): Finally, x i is used as the input to the model.”
InfluentialDeep Learning Approach to Recognize Yoga Posture for the Ailment of the Low Back Pain
Kiran, K. U., Singh, M., Ansari, M. S. A., & Govil, M. C. (2023) · Lecture notes in electrical engineering · doi:10.1007/978-981-99-2710-4_21 · Cited by 5
Ye, N., Zhang, L., Xiong, D., Wu, H., & Song, A. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3555985 · Cited by 7
“Ordóñez and Roggen [23] combined long short-term memory networks (LSTMs) with CNNs, demonstrating unique advantages in multimodal sensor signal recognition tasks.”
Calorific Expenditure Estimation Using Deep Convolutional Network Features
Wang, B., Tao, L., Burghardt, T., & Mirmehdi, M. (2018) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw.2018.00014 · Cited by 8
“There are a few recent studies on healthcare related applications using deep learning framework [9, 21], however to the best of our knowledge, our work is the first one targeted at improving the estimation of energy expenditure using visual data only utilising deep neural networks.”
BackgroundSmartphone-Based Activity Recognition in a Pedestrian Navigation Context
Jackermeier, R., & Ludwig, B. (2021) · Sensors · doi:10.3390/s21093243 · Cited by 6
“As it is one of the most consistent and well-performing approaches in their comparative evaluation, the DeepConvLSTM method [10] serves as a baseline in our evaluation later on.”
MethodChandrasekaran, K., Gerych, W., Buquicchio, L., Alajaji, A. S., Agu, E., & Rundensteiner, E. A. (2021) · IEEE International Conference on Smart Computing · doi:10.1109/smartcomp52413.2021.00026 · Cited by 4
Alfayez, F., & Bouhamed, H. (2023) · International Journal of Computers Communications & Control · doi:10.15837/ijccc.2023.1.4981 · Cited by 4
A customized 1D-CNN approach for sensor-based human activity recognition
(2022) · International Journal of Advanced Technology and Engineering Exploration · doi:10.19101/ijatee.2021.874828 · Cited by 7
Long kernel distillation in human activity recognition
Yao, M., Cheng, D., Zhang, L., Li, W., Wu, H., & Song, A. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113397 · Cited by 6
Deep domain adaptation eliminates costly data required for task-agnostic wearable robotic control
Scherpereel, K. L., Gombolay, M., Shepherd, M. K., Carrasquillo, C., Inan, O. T., & Young, A. J. (2025) · Science Robotics · doi:10.1126/scirobotics.ads8652 · Cited by 11
GrCAN: Gradient Boost Convolutional Autoencoder with Neural Decision ForestPreprint
Dong, M., Yao, L., Wang, X., Benatallah, B., & Zhang, S. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.08079 · Cited by 14
“Especially in recent years, they got lots of achievements in dealing with image recognition [36], activity recognition [2], document classification [13], and many other new tasks [7].”
BackgroundElahi, S., Mery, T., Panthier, C., Saad, A., Gatinel, D., & Alahi, A. (2024) · Translational Vision Science & Technology · doi:10.1167/tvst.13.9.1 · Cited by 6
“29 In recent years, the application of transformer models in multivariate time series classification tasks, including human activity recognition, has been explored.”
BackgroundMethodMarhraoui, Y. E., Amroun, H., Boukallel, M., Anastassova, M., Lamy-Perbal, S., Bouilland, S., & Ammi, M. (2022) · Computers · doi:10.3390/computers11050058 · Cited by 7
“[53] used this type of layer to conduct their study of activity recognition using as well different sources of inputs; we configured our model to take into account the data format we are adopting in this case.”
Methodde Gauna, D. E. R., Sanchez, L. E., & Ruiz‐Iniesta, A. (2023) · PeerJ Computer Science · doi:10.7717/peerj-cs.1534 · Cited by 4
“Some approaches apply Multi-step Dense to classi fi cation problems by training them in a supervised manner; LSTM-RNN in time series problems; and CNNs in order to learn multiple layers of feature hierarchies automatically ( Ordóñez & Roggen, 2016 ).”
MethodWasim, M., Ahmed, I., Abbas, N., Saba, T., Alamri, F. S., Elyassih, A., & Rehman, A. U. (2025) · Scientific Reports · doi:10.1038/s41598-025-07620-3 · Cited by 9
“Ordonez at el. have proposed a deep recurrent neural network 21 with a combination of a sequence of convolutional and LSTM layers for long activity sequences.”
Research on Data Cleaning Algorithm Based on Multi Type Construction Waste
Wang, P., Liu, Y., Sun, Q., Bai, Y., & Li, C. (2022) · Sustainability · doi:10.3390/su141912286 · Cited by 4
A Many-Objective Simultaneous Feature Selection and Discretization for LCS-Based Gesture Recognition
Otis, M. J.-D., & Vandewynckel, J. (2021) · Applied Sciences · doi:10.3390/app11219787 · Cited by 7
GA-LSTM-Based Degradation Prediction for IGBTs in Power Electronic Systems
Qiu, Y., Li, Z., & Tian, S. (2025) · Energies · doi:10.3390/en18215574 · Cited by 5
Classifying Sport-Related Human Activity from Thermal Vision Sensors Using CNN and LSTM
Polo-Rodríguez, A., Lendínez, A. M., Espinilla, M., & Medina-Quero, J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-13321-3_4 · Cited by 5
Improving Human Activity Recognition Models by Learnable Sparse Wavelet Layer
Zhao, H., Zhou, Y., Riedel, T., Hefenbrock, M., & Beigl, M. (2022) · ACM International Symposium on Wearable Computers (ISWC) · doi:10.1145/3544794.3558461 · Cited by 4
“Toverifytheeffectivenessofthe learnablesparsewavelet layer , we apply the layer to three SOTA benchmark HAR models, DeepConvLSTM [21], Multibranch CNN (MCNN) [20] and Self-Attention HAR (SA-HAR) [17].”
MethodSegment-Based Unsupervised Learning Method in Sensor-Based Human Activity Recognition
Takenaka, K., Kondo, K., & Hasegawa, T. (2023) · Sensors · doi:10.3390/s23208449 · Cited by 5
Enhancing Efficiency in HAR Models: NAS Meets Pruning
Zhou, Y., King, T., Huang, Y., Zhao, H., Riedel, T., Röddiger, T., & Beigl, M. (2024) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops59983.2024.10502894 · Cited by 6
Discriminative multiscale CNN network for smartphone based robust gait recognition
Das, S., Meher, S., & Sahoo, U. K. (2021) · Indian Conference on Computer Vision, Graphics and Image Processing · doi:10.1145/3490035.3490308 · Cited by 6
A Recurrent Neural Network Optimization Method for Anticipation of Hierarchical Human Activity
Khodabandelou, G., Chibani, A., Amirat, Y., & Tanimoto, S. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2023.3300821 · Cited by 4
“Following [48], 113 features and all 5 activities are selected.”
BackgroundApplication of 2-D Convolutional Neural Networks for Damage Detection in Steel Frame StructuresPreprint
Ghazvineh, S., Nouri, G., Lavassani, S. H. H., Gharehbaghi, V., & Nguyễn, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.15895 · Cited by 11
Hi-OSCAR: Hierarchical Open-set Classifier for Human Activity Recognition
McCarthy, C., Quirijnen, L., van Zandwijk, J. P., Geradts, Z., & Worring, M. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3770681 · Cited by 6
“Examples are CNN layers combined with LSTMs, BiLSTMs, GRUs, and BiGRUs [60, 51] which are effective for classification, as well as other variations incorporating residual modules [32] and unique combinations of CNNs based on modality before combination and being fed into an LSTM [18].”
A Review of Animal Individual Recognition Based on Computer Vision
Wang, W., Wu, J., Yu, H. Z., Zhang, H., Zhou, Y., & Zhang, Y.-X. (2022) · Communications in computer and information science · doi:10.1007/978-981-19-5194-7_22 · Cited by 5
Xia, W., Xu, J., Liu, B., & Duan, H. (2024) · Energy Science & Engineering · doi:10.1002/ese3.1823 · Cited by 7
Exploiting Video Classification Using Deep Learning Models for Human Activity Recognition
Singh, U., & Singhal, N. (2023) · Algorithms for intelligent systems · doi:10.1007/978-981-19-7892-0_14 · Cited by 4
Stationary Exercise Classification using IMUs and Deep Learning
Heroy, A., Gill, Z., Sprague, S., Stroud, D., & Santerre, J. (2020) · SMU Data Science Review · Cited by 4
Wi-Exercise: An Indoor Human Movement Detection Method Based on Bidirectional LSTM Attention
Hao, Z., Kang, Y., Dang, X., & An, Y. (2022) · Mobile Information Systems · doi:10.1155/2022/9994304 · Cited by 4
Exploring System Performance of Continual Learning for Mobile and Embedded Sensing Applications
Kwon, Y. D., Chauhan, J., Kumar, A., Hui, P., & Mascolo, C. (2021) · ePrints Soton (University of Southampton) · doi:10.17863/cam.81436 · Cited by 26
Dhar, J., Rana, K., Goyal, P., Alavi, A., Rana, R. K., Vo, Q. B., Mishra, S., & Mistry, S. (2024) · Applied Soft Computing · doi:10.1016/j.asoc.2024.112550 · Cited by 5
Tello, A., & Degeler, V. (2025) · ACM Symposium on Applied Computing · doi:10.1145/3672608.3707743 · Cited by 5
“Namely, CNN [32], 2-LSTM [23], 4-CNN-LSTM [23], DeepConvLSTM [26], Self-Attention [22], and ResGCNN [37].”
Slow Feature Preprocessing in Deep Neural Networks for Wearable Sensor-Based Locomotion Recognition
Günthermann, L., Pellatt, L., & Roggen, D. (2022) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops53856.2022.9767511 · Cited by 4
“In [11], the kernel size for this dataset is chosen as 5 samples (approximately 0.16 seconds), and the window size is varied between 0.5 and 4 seconds.”
BackgroundMethodInfluentialSelf-citationKanoga, S., Tsukiji, Y., & Karakida, R. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.127469 · Cited by 5
Yao, R., Lin, G., Shi, Q., et al. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1702.06212 · Cited by 4
Long-term self-supervised learning for accelerometer-based sleep–wake recognition
Logacjov, A., Bach, K., & Mork, P. J. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109758 · Cited by 8
Pluto: Motion Detection for Navigation in a VR Headset
Kovalenko, D., Migukin, A., Ryabkova, S., & Chernov, V. (2021) · International Conference on Indoor Positioning and Indoor Navigation · doi:10.1109/ipin51156.2021.9662466 · Cited by 4
“Attempts to recognize pedestrian action were conducted with various setups: from a minimalistic wristband [22] to a set of 19 IMUs mounted on every part of a human body [23], with 95% F1 score reported by latter with Deep Convolutional LSTM for 17 actions classification.”
MethodPredicting Adolescent Female Stress with Wearable Device Data Using Machine and Deep Learning
Jin, C., Osotsi, A., & Oravecz, Z. (2023) · IEEE International Conference on Body Sensor Networks (BSN) · doi:10.1109/bsn58485.2023.10331414 · Cited by 4
Hiremath, S. K., & Ploetz, T. (2021) · 2021 International Symposium on Wearable Computers · doi:10.1145/3460421.3478825 · Cited by 4
“The de-facto standard here is the DeepConvLSTM architecture [25] that combines four convolutional layers with 64 learned filters (dimensionality: 5x1) for hierarchical feature extraction, with two Long Short Term Memory (LSTM [15]) layers to capture sequential information.”
BackgroundMethodInfluentialOstovan, M., Samadi, S., & Kazemi, A. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/7365544 · Cited by 4
Deep-learning-based unobtrusive handedness prediction for one-handed smartphone interaction
Chen, T., Zhu, K., & Yang, M. C. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11844-6 · Cited by 4
Shi, Y., & Watada, J. (2015) · ICIC Express Letters · Cited by 4
A Trio Neural Model for Dynamic Entity Relatedness Ranking
Nguyen, T. N., Tran, T., & Nejdl, W. (2018) · Conference on Computational Natural Language Learning · doi:10.18653/v1/k18-1004 · Cited by 4
Depthwise-Separable U-Net for Wearable Sensor-Based Human Activity Recognition
Lee, Y.-K., Son, C.-S., & Kang, W. (2025) · Applied Sciences · doi:10.3390/app15169134 · Cited by 4
Mekruksavanich, S., & Jitpattanakul, A. (2025) · Informatics · doi:10.3390/informatics12040115 · Cited by 4
“This dual capability allows for a comprehensive analysis of human activity data’s spatial and temporal dimensions [34].”
BodyFlow: An Open-Source Library for Multimodal Human Activity Recognition
Alonso, R. D. H., Hernández-Ruiz, A. C., Marañes, C., López-Bosque, I., Aznar-Gimeno, R., Salvo-Ibáñez, P., Pérez-Lázaro, P., Abadía, D., & Rodrigálvarez-Chamarro, M. D. L. V. (2024) · Sensors · doi:10.3390/s24206729 · Cited by 4
“…choice in this context, particularly due to their ability to automate feature extraction from raw sensor inputs while modeling temporal dependencies, as demonstrated by Ordóñez and Roggen’s work on combining deep convolutional and LSTM recurrent networks for human activity recognition [27].”
BackgroundDouble Doodles: Sketching Animation in Immersive Environment With 3+6 DOFs Motion Gestures
Chen, R., Pan, Y., Deng, Z., Wang, L., & Ma, L. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3613783 · Cited by 5
“It uses the DeepConvLSTM structure [18] as the basic structure.”
MethodLago, P., & Inoue, S. (2019) · International Conference on Informatics, Electronics & Vision (ICIEV) · doi:10.1109/iciev.2019.8858548 · Cited by 4
“To better understand these results, we further explore the average precision and recall for each class on the PCA (best performance) and LSTM (lowest performance) methods in Figure 4.”
MethodInfluentialFold bifurcation identification through scientific machine learning
Habib, G., & Horváth, Á. (2024) · Physica D Nonlinear Phenomena · doi:10.1016/j.physd.2024.134490 · Cited by 5
“Examples include but are not limited to, human activity recognition [47, 48], electrocardiogram classification [49, 50], speech recognition [51, 52], natural language processing [53, 54].”
Deep Learning Algorithms for Human Activity Recognition in Manual Material Handling Tasks
Bassani, G., Avizzano, C. A., & Filippeschi, A. (2025) · Sensors · doi:10.3390/s25216705 · Cited by 7
“Finally, we compare the selected neural networks with the DeepConvLSTM [35], a recurrent network already used as a benchmark [36], in terms of recognition ability and computational complexity.”
InfluentialCHAR: Composite Head-Body Activities Recognition With a Single Earable Device
Zhu, P., Zhu, Y., Li, W., He, Y., Zou, Y., Wu, K., & Leung, V. C. M. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3548647 · Cited by 8
“In this section, we compare our proposed network CARN with several classical networks for HAR as benchmarks including DCNN [22], DeepSense [23], and DeepConvLSTM [15].”
InfluentialLi, K., Tian, S., Yu, L., Zhou, T., Wang, B., & Wang, F. (2023) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-222189 · Cited by 4
“A variety of effective fusion methods have been proposed, such as tensor-based fusion and attention-based models [9, 23, 26, 40].”
Hybrid Neural Network Methods for the Detection of Credit Card Fraud
Al‐Khasawneh, M. A., Faheem, M., Alsekait, D. M., Abubakar, A., & Issa, G. F. (2025) · Security and Privacy · doi:10.1002/spy2.500 · Cited by 5
“Ordonez and Roggen [16] proposed a multi-modal Recurrent Neural Network (m-RNN) model.”
Wang, L., Yan, Y., Huang, D., Pan, Y., Cheang, C. F., Luo, K., & Li, J. (2025) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2025.3540853 · Cited by 6
“To comprehensively compare the performance of the AHARNet, we compared it with various structural models, such as LSTM [35], [36], CNN [19], [34], ConvLSTM [20], Attention [37], [38], [39], and Transformer [22].”
SemNet: Learning semantic attributes for human activity recognition with deep belief networks
Venkatachalam, S., Nair, H., Zeng, M., Tan, C. S., Mengshoel, O. J., & Shen, J. P. (2022) · Frontiers in Big Data · doi:10.3389/fdata.2022.879389 · Cited by 4
“Many works successfully combine CNNs and RNNs to perform complex activity recognition (Ordóñez and Roggen, 2016; Zhao et al., 2018; Xu et al., 2019).”
BackgroundDaily life human activities recognition using a novel AR‐DenseNet
Ghalan, M., & Aggarwal, R. K. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.7142 · Cited by 4
Sipos, D., Vészi, K., Bogár, B., Pető, D., Füredi, G., Betlehem, J., & Pandur, A. A. (2025) · Sci · doi:10.3390/sci7040137 · Cited by 8
“Miniaturized, low-power biosensors provide continuous, accurate physiological data streams, while hierarchical deep learning frameworks rapidly analyze these data to predict injury risk and physiological strain with impressive accuracy [27–31].”
Human Activity Recognition Based on Hybrid Learning Algorithm for Wearable Sensor DataPreprint
Athota, R. K., & Sumathi, D. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4162745 · Cited by 4
Cross-Domain Activity Recognition Using Shared Representation in Sensor Data
Hamad, R. A., Yang, L., Woo, W. L., & Wei, B. W. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3178083 · Cited by 6
“A deep convolutional LSTM is proposed for multimodal wearable activity recognition [23].”
MethodYe, N., Zhang, L., Xiong, D., Wu, H., & Song, A. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3458441 · Cited by 5
“Ordóñez et al. [26] have introduced a powerful architecture termed DeepConvLSTM via embedding LSTM units into standard CNNs, which presents the unique advantage of recurrent structures in extracting temporal features from multimodal sensor signals.”
Stacked LSTM for Short-Term Wind Power Forecasting Using Multivariate Time Series Data.
Galphade, M., Nikam, V. B., Banerjee, B., et al. (2024) · International Journal of Interactive Multimedia and Artificial Intelligence · doi:10.9781/ijimai.2024.07.002 · Cited by 4
Rey, V. F., Rezende, P. M. B., Zhou, B., Suh, S., & Lukowicz, P. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129288 · Cited by 14
Asymmetric Loss Functions for Contract Capacity Optimization
Lin, J.-L., Zhang, Y., Zhu, K., Chen, B., & Zhang, F. (2020) · Energies · doi:10.3390/en13123123 · Cited by 4
Lange, M., Kirkham, R., & Tannert, B. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-85616-8_25 · Cited by 6
Model-Agnostic Structural Transfer Learning for Cross-Domain Autonomous Activity Recognition
Alinia, P., Arefeen, A., Ashari, Z. E., Mirzadeh, S. I., & Ghasemzadeh, H. G. (2023) · Sensors · doi:10.3390/s23146337 · Cited by 5
Deep convolutional state space model as human activity recognizer
Wang, L., Bu, C., Yao, M., Xiong, D., Wang, S., Cheng, D., Zhang, L., Wu, H., & Song, A. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103982 · Cited by 8
Mekruksavanich, S., Hnoohom, N., Phaphan, W., & Jitpattanakul, A. (2025) · Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering (ECTI DAMT & NCON) · doi:10.1109/ectidamtncon64748.2025.10962074 · Cited by 6
Eff-WHAR: A Lightweight Design for Efficient Wearable Sensor-Based Human Activity Recognition
Chen, Z., Zou, H., Wang, L., Wang, B., Zhang, F., Ma, S., Pan, Y., & Li, J. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3509961 · Cited by 8
Multi-level Contrast Network for Wearables-based Joint Activity Segmentation and Recognition
Xia, S., Chu, L., Pei, L., Yu, W., & Qiu, R. C. (2022) · Globecom · doi:10.1109/globecom48099.2022.10001226 · Cited by 5
“The researchers utilized deep neural networks(DNNs) [11], convolutional neural networks(CNNs) [12], recurrent neural networks(RNNs) [13] or hybrid structures [14] to extract features from the related time series and then effectively recognized corresponding activities.”
BackgroundMethodZhao, T., Chen, H., Bai, Y., Zhao, Y., & Zhao, S. (2022) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph191811706 · Cited by 4
Activity-Free User Identification Using Wearables Based on Vision Techniques
Guinea, A. S., Heinrich, S., & Mühlhäuser, M. (2022) · Sensors · doi:10.3390/s22197368 · Cited by 5
Lim, H., & Myung, H. (2019) · Communications in computer and information science · doi:10.1007/978-981-13-7780-8_12 · Cited by 4
High-Intensity Interval Training Exercise Recognition using Smartwatch
Sivakumar, S., Kun, Y. J., & Gopalai, A. A. (2021) · IEEE-EMBS Conference on Biomedical Engineering and Sciences · doi:10.1109/iecbes48179.2021.9398735 · Cited by 4
“Firstly the hyper-parameters were tuned using the Grid Search method [7], [10], [18] and followed by the Bayesian Optimization method [19].”
MethodRepresenting Temporal Dependencies in Smart Home Activity Recognition for Health Monitoring
Forbes, G., Massie, S., Craw, S., Fraser, L., & Hamilton, G. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9207480 · Cited by 5
“This may be explicitly by employing a feature engineering approach, however, manually chosen features using heuristic processes may be outperformed by modern deep learning research [11].”
MethodHybrid deep neural networks to infer state models of black-box systems
Mashhadi, M. J., & Hemmati, H. (2020) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1145/3324884.3416559 · Cited by 9
“ned in one single convolutional layer. Recurrentneuralnetworks(RNN)haveshowngreatperformance in analysing sequential data such as machine translation, timeseriesprediction,andtime-seriesclassification[16,54,56,79,82,86]. RNNs can capture long-term temporal dependencies which is quite useful for solving our problem. [1…”
BackgroundResultMekruksavanich, S., Jantawong, P., & Jitpattanakul, A. (2024) · International Conference on Big Data Analytics and Practices (IBDAP) · doi:10.1109/ibdap62940.2024.10689690 · Cited by 5
“This method involves using a CNN to identify distinct spatial characteristics from unprocessed sensor input and an LSTM to record extended temporal relationships within the sequence of features [29].”
MethodActivity recognition on smartphones using an AKNN based support vectors
Abidine, M. B., Oussalah, M., Fergani, B., & Lounis, H. (2022) · Sensor Review · doi:10.1108/sr-05-2021-0157 · Cited by 4
Impact of Three-Dimensional Video Scalability on Multi-View Activity Recognition using Deep Learning
Choi, J. H., Choi, M., Choi, M., & Lee, J. (2017) · Thematic Workshops of ACM Multimedia · doi:10.1145/3126686.3126769 · Cited by 5
“Our model is similar to the long-term recurrent convolutional network (LRCN) in [5] and DeepConvLSTM in [16].”
MethodHiding in the deep : online animal activity recognition using motion sensors and machine learning
Kamminga, J. W. (2020) · doi:10.3990/1.9789036550550 · Cited by 5
Kamikokuryo, K., Venture, G., & Hernandez, V. (2025) · Smart Health · doi:10.1016/j.smhl.2024.100537 · Cited by 5
Video-based Pose-Estimation Data as Source for Transfer Learning in Human Activity Recognition
Awasthi, S., Rueda, F. M., & Fink, G. A. (2022) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr56361.2022.9956405 · Cited by 5
“In the case of having [10, 30, 50]% of the Dt, the pre-trained networks on JHMDB, CAD60 and Penn significantly improved the performance.”
BackgroundMethodInfluentialEffect of Combinations of Sensor Positions on Wearable-sensor-based Human Activity Recognition
Duan, Y., & Fujinami, K. (2023) · Sensors and Materials · doi:10.18494/sam4421 · Cited by 4
Lin, T.-J. C., Hsu, C.-W., Lee, Y., Shieh, T.-H., & Wang, Y.-H. (2025) · International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) · doi:10.1109/cyberc66434.2025.00027 · Cited by 6
“For time series data, the hybrid approach of CNN and LSTM has been effective in research fields such as sensor data analysis [26].”
Human Activity Recognition (HAR) in Healthcare, 2nd Edition
Bibbò, L., & Serrano, A. (2025) · Applied Sciences · doi:10.3390/app15105762 · Cited by 5
Temporal Multimodal Fusion for Driver Behavior Prediction Tasks using Gated Recurrent Fusion Units.
Narayanan, A., Siravuru, A., & Dariush, B. (2019) · Cited by 8
CHAR: Composite Head-body Activities Recognition with A Single Earable Device
Zhu, P., Zou, Y., Li, W., & Wu, K. (2023) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom56429.2023.10099218 · Cited by 6
“In this section, we compare our proposed network CARN with several classical networks for HAR as benchmarks including DCNN [22], DeepSense [23], and DeepConvLSTM [15].”
BackgroundMethodInfluentialBusran, N. M., Putrada, A. G., & Wicaksono, R. L. (2025) · International Conference on Data Science and Its Applications (ICoDSA) · doi:10.1109/icodsa67155.2025.11157039 · Cited by 4
Chen, T., Liu, X., Xia, B., et al. (2020) · IEEE Access · doi:10.1109/access.2020.3004098 · Cited by 4
Heterogeneous Non-Local Fusion for Multimodal Activity Recognition
Byvshev, P., Mettes, P., & Xiao, Y. (2020) · International Conference on Multimedia Retrieval · doi:10.1145/3372278.3390675 · Cited by 4
“been studied using late convolutional fusion followed by fullyconnected layers [31], using convolutional LSTMs on top of sensory-independent representations [44], or through multiobjective regularization on different embedding combinations”
MethodSurvey on Deep Learning Based Fusion Recognition of Multimodal Biometrics
Yang, Q., Chen, X., He, Z., & Chang, L. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20233-9_52 · Cited by 7
Node-level parallelization for deep neural networks with conditional independent graph
Zhou, F., Wu, F., Zhang, Z., & Dong, M. (2017) · Neurocomputing · doi:10.1016/j.neucom.2017.06.002 · Cited by 4
Kianpisheh, M., Mariakakis, A. T., & Truong, K. N. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3643500 · Cited by 8
Design of a CNN based autonomous in-seat passenger anomaly detection system
Bhatnagar, A., Nandi, P., Mishra, A., et al. (2022) · 2020 IEEE International Systems Conference (SysCon) · doi:10.1109/syscon53536.2022.9773868 · Cited by 4
A method of classification decision based on multi-BiLSTMs for physical loads hierarchy
Wang, Y., Zhang, C., Zhao, Y., Liao, Y., Gao, Y., & Zheng, J. (2022) · Computer Methods in Biomechanics & Biomedical Engineering · doi:10.1080/10255842.2022.2106785 · Cited by 4
“…(14) Precision ¼ TP TP þ FP (15) F score ¼ 2 Precision Sensitivity Precision þ Sensitivity (16) Accuracy ¼ TP þ TN TP þ FP þ TN þ FN (17) In the classification problem (Ord o~nez and Roggen 2016), the F-score is used to measure the performance of the classification model by avoiding the biased…”
MethodHybrid spiking neural model for clustering smart environment activities
Amin, H. H., Deabes, W. A., & Bouazza, K. (2017) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin.2017.8104772 · Cited by 4
“These models have limited performances due to many reasons such as the use of only sensors reading values regardless the time of these readings [8]–[10].”
BackgroundMulti-attention deep recurrent neural network for nursing action evaluation using wearable sensor
Zhong, Z., Lin, C., Ogata, T., & Ota, J. (2020) · ACM International Conference on Intelligent User Interfaces (IUI) · doi:10.1145/3377325.3377530 · Cited by 4
“Thereafter, owing to the better performance of recurrent neural networks (RNNs) in the time-series domain, several variants of RNNs, such as multicolumnBiRNN [29], DeepConvLSTM [23], ResBidirLSTM [34], and DRNN [21] have been proposed, which represent the state-of-theart level wearable sensor-based HAR.”
Background
Metadata
- Authors
- Ordóñez, F., & Roggen, D.
- Affiliations
- University of Sussex (GB)
- Corresponding author
- Ordóñez, Francisco
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2016-01-18
- Volume
- 16
- Issue
- 1
- Pages
- 115
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC4732148
- OpenAlex
- W2270470215
- Cited by
- 2,761
- References
- 49
- Citations by year
- 2026: 284 · 2025: 256 · 2024: 284 · 2023: 270 · 2022: 319 · 2021: 379 · 2020: 347 · 2019: 288 · 2018: 219 · 2017: 93 · 2016: 21
- FWCI
- 140.52
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Sussex Research Online (University of Sussex) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by)
- Funders
- Google Faculty Research Award (Is deep learning useful for wearable activity recognition?)
- EPSRC (EP/N007816/1)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Engineering › Biomedical Engineering › Gait Recognition and Analysis
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Keywords
- human activity recognition, wearable sensors, multimodal sensor fusion, convolutional neural networks, LSTM neural network, temporal dynamics, automatic feature extraction, hyperparameter optimization
- MeSH terms
- Machine Learning, Clothing, Humans, Human Activities, Signal Processing, Computer-Assisted, Databases, Factual, Neural Networks, Computer, Monitoring, Ambulatory
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.