Improving Cross-Subject Activity Recognition via Adversarial Learning
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
Deep learning has been widely used for implementing human activity recognition from wearable sensors like inertial measurement units. The performance of deep activity recognition is heavily affected by the amount and variability of the labeled data available for training the deep learning models. On the other hand, it is costly and time-consuming to collect and label data. Given limited training data, it is hard to maintain high performance across a wide range of subjects, due to the differences in the underlying data distribution of the training and the testing sets. In this work, we develop a novel solution that applies adversarial learning to improve cross-subject performance by generating training data that mimic artificial subjects - i.e. through data augmentation - and enforcing the activity classifier to ignore subject-dependent information. Contrary to domain adaptation methods, our solution does not utilize any data from subjects of the test set (or target domain). Furthermore, our solution is versatile as it can be utilized together with any deep neural network as the classifier. Considering the open dataset PAMAP2, nearly 10% higher cross-subject performance in terms of F1-score can be achieved when training a CNN-LSTM-based classifier with our solution. A performance gain of 5% is also observed when our solution is applied to a state-of-the-art HAR classifier composed of a combination of inception neural network and recurrent neural network. We also investigate different influencing factors of classification performance (i.e. selection of sensor modalities, sampling rates and the number of subjects in the training data), and summarize a practical guideline for implementing deep learning solutions for sensor-based human activity recognition.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761 · cited 5×
“The performances for DeepConvLSTM [26] and the LSTM with Uniqueness Attention [41] are also reported.”
BackgroundMethodInfluentialDeep visual domain adaptation: A survey
Wang, M., & Deng, W. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.05.083 · Cited by 2,209
Introducing a New Benchmarked Dataset for Activity Monitoring
Reiss, A., & Stricker, D. (2012) · 2012 16th International Symposium on Wearable Computers · doi:10.1109/iswc.2012.13 · Cited by 1,456 · cited 17×
“Their experiments were conducted with public datasets as PAMAP2.”
BackgroundMethodInfluentialThe Opportunity challenge: A benchmark database for on-body sensor-based activity recognition
Chavarriaga, R., Sagha, H., Calatroni, A., Digumarti, S. T., Tröster, G., Millán, J. D. R., & Roggen, D. (2013) · Pattern Recognition Letters · doi:10.1016/j.patrec.2012.12.014 · Cited by 742 · cited 1×
“Opportunity [3] and Daphnet [1], typically contain no more than 10 subjects.”
BackgroundWearable Assistant for Parkinson’s Disease Patients With the Freezing of Gait Symptom
Bachlin, M., Plotnik, M., Roggen, D., Maidan, I., Hausdorff, J. M., Giladi, N., & Troster, G. (2010) · IEEE Transactions on Information Technology in Biomedicine · doi:10.1109/titb.2009.2036165 · Cited by 693 · cited 10×
“Opportunity [3] and Daphnet [1], typically contain no more than 10 subjects.”
BackgroundMethodInfluentialDeep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables
Hammerla, N. Y., Halloran, S., & Ploetz, T. (2016) · arXiv · doi:10.48550/arxiv.1604.08880 · Cited by 606
Ensembles of Deep LSTM Learners for Activity Recognition using Wearables
Guan, Y., & Plötz, T. (2017) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3090076 · Cited by 511 · cited 18×
“The performances for DeepConvLSTM [26] and the LSTM with Uniqueness Attention [41] are also reported.”
BackgroundMethodInfluentialInnoHAR: A Deep Neural Network for Complex Human Activity Recognition
Xu, C., Chai, D., He, J., Zhang, X., & Duan, S. (2019) · IEEE Access · doi:10.1109/access.2018.2890675 · Cited by 333 · cited 3×
“Additionally to having a CNN-LSTM network as the activity classifier, we also performed tests having InnoHAR [36] - composed of inception layers followed by GRU layers - as the activity classifier since it has exhibited state-of-the-art performance in HAR.”
MethodInfluentialZeng, M., Gao, H., Yu, T., Mengshoel, O. J., Langseth, H., Lane, I., & Liu, X. (2018) · Proceedings of the 2018 ACM International Symposium on Wearable Computers · doi:10.1145/3267242.3267286 · Cited by 185 · cited 3×
“This is a common choice of subjects in the literature [7], [8], [26], [37], [40].”
BackgroundMethodPeng (2018) · doi:10.1145/3214277 · Cited by 134
Research works printed but not linked: 29
- 10.1109/PERCOM.2018.8444594 · DOI
- 10.1145/3265689.3265705 · DOI
- 10.1109/CVPR.2017.316 · DOI · Paper page
- springenberg, 2015, Unsupervised and semi-supervised learning with categorical generative adversarial networks, CoRR, abs 1511 6390, 1
- 10.1109/ACCESS.2018.2873315 · DOI
- 10.1109/JSAC.2017.2679658 · DOI · Paper page
- 10.1145/3241539.3241548 · DOI · Paper page
- karras, 2017, Progressive growing of Gans for improved quality, stability, and variation, CoRR, abs 1710 10196, 1
- karras, 2018, A style-based generator architecture for generative adversarial networks, CoRR, abs 1812 4948, 1
- 10.1109/ISCAS.2018.8351727 · doi:10.1109/iscas.2018.8351727
- 10.1109/PERCOM.2018.8444585 · DOI · Paper page
- 10.1109/SMC.2017.8123083 · doi:10.1109/smc.2017.8123083
- long, 2019, Dual residual network for accurate human activity recognition, CoRR, abs 1903 5359, 1
- miyato, 2018, Spectral normalization for generative adversarial networks, CoRR, abs 1802 5957, 1
- 10.1145/3152832.3152844 · DOI
- 10.1016/j.aei.2019.100944 · DOI · Paper page
- 10.1007/978-3-319-58347-1_1 · DOI · Paper page
- goodfellow, 2014, Generative adversarial nets, Proc Adv Neural Inf Process Syst, 2672
- 10.1109/RADAR.2019.8835589 · DOI
- barbosa, 2018, Unsupervised Domain Adaptation for Human Activity Recognition, 623
- 10.1109/TII.2017.2779814 · DOI · Paper page
- 10.1109/ISCAS.2018.8351613 · DOI · Paper page
- 10.1145/3123021.3123046 · DOI · Paper page
- 10.1145/3134230.3134236 · DOI
- 10.1109/ICCV.2017.244 · DOI · Paper page
- 10.1109/BSN.2018.8329677 · DOI
- 10.1007/978-3-030-30508-6_40 · DOI
- 10.1145/2638728.2641304 · DOI
- 10.1145/3123024.3123095 · DOI
Cited by
Suh, S., Rey, V. F., & Lukowicz, P. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.110143 · Cited by 107
IMUGPT 2.0: Language-Based Cross Modality Transfer for Sensor-Based Human Activity Recognition
Leng, Z., Bhattacharjee, A., Rajasekhar, H., Zhang, L., Bruda, E., Kwon, H., & Plötz, T. (2024) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3678545 · Cited by 61
“Leng et al. [43] piloted the concept of IMUGPT that combines large language models (LLMs), motion synthesis methods, and signal processing techniques to generate virtual IMU data with no manual effort.”
Haresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Plötz, T. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3729467 · Cited by 33
“These methods include jittering, scaling, rotation, and random sampling, which have demonstrated efficacy in enhancing the diversity of the training dataset.”
If only we had more data!: Sensor-Based Human Activity Recognition in Challenging Scenarios
Plötz, T. (2023) · 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) · doi:10.1109/percomworkshops56833.2023.10150267 · Cited by 16
“Another trend is to combine real as well as virtual data with simulated and/or augmented samples [70].”
BackgroundHaresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Ploetz, T. (2024) · arXiv · Cited by 31
“These methods include jittering, scaling, rotation, and random sampling, which have demonstrated efficacy in enhancing the diversity of the training dataset.”
Metadata
- Authors
- Leite, C. F. S., & Xiao, Y.
- Venue
- IEEE Access
- Published
- 2020
- Volume
- 8
- Pages
- 90542-90554
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W3025821620
- Cited by
- 18
- References
- 13
- FWCI
- 0.88
- Type
- journal article
- Language
- en
- Open access
- gold
- Funders
- Business Finland (1660/31/2018)
- European Union’s Horizon 2020 Research and Innovation Programme (777222)
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.
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Improving Cross-Subject Activity Recognition via Adversarial Learning. https://app.cochat.ai/papers/10.1109/access.2020.2993818
- The paper
- Leite, C. F. S., & Xiao, Y. (2020). Improving Cross-Subject Activity Recognition via Adversarial Learning. IEEE Access. https://doi.org/10.1109/access.2020.2993818
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.