Models

Model catalog for the static WHAR Arena paper snapshot.

Snapshot: May 1, 2026
Models: 17

17 Models

Name		Architecture
TinierHAR German Research Center for Artificial Intelligence (DFKI)	2025	Attention ⓘ CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	7.3k	892.3k
MLP-HAR Karlsruhe Institute of Technology	2024	Dense ⓘ Spectral ⓘ	PyTorch	157.7k	2.3M
MLP-Mixer University of Southampton	2023	Dense ⓘ	PyTorch	605.9k	87.5M
DynamicWHAR Zhejiang University	2022	CNN ⓘ Dense ⓘ Graph ⓘ	PyTorch	1.12M	14.7M
TinyHAR Karlsruhe Institute of Technology	2022	Attention ⓘ CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	159.1k	9.3M
Triple-Cross-Attn Nanjing Normal University, Southeast University	2022	Attention ⓘ CNN ⓘ Dense ⓘ	PyTorch	278.3k	17.9M
Attend+Discriminate The University of Adelaide	2021	Attention ⓘ CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	371.5k	171.5M
DANA Imperial College London, Queen Mary University of London	2021	CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	457.9k	102.4M
DeepConvShallowLSTM University of Siegen	2021	CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	399.9k	207.6M
SA-HAR University of Dhaka, Independent University Bangladesh	2020	Attention ⓘ CNN ⓘ Dense ⓘ	PyTorch	401.3k	121.5M
DeepConvLSTM-Attn Georgia Institute of Technology	2018	Attention ⓘ CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	474.6k	179.9M
Guan-LSTM Newcastle University, Georgia Institute of Technology	2017	Dense ⓘ RNN ⓘ	PyTorch	798.0k	205.6M
DeepConvLSTM University of Sussex	2016	CNN ⓘ Dense ⓘ RNN ⓘ	PyTorch	457.9k	176.2M
CNN-HAR A*STAR	2015	CNN ⓘ	PyTorch	63.2k	11.3M
k-NN Baseline	-	Features ⓘ	scikit-learn	-	-
Random Forest Baseline	-	Features ⓘ	scikit-learn	-	-
SVM Baseline	-	Features ⓘ	scikit-learn	-	-

Citation

@misc{burzer2026whararenabenchmarkingstate,
  title = {{WHAR Arena}: Benchmarking the State of the Art in Efficient Wearable Human Activity Recognition},
  author = {Maximilian Burzer and Tobias King and Till Riedel and Michael Beigl and Tobias R{"o}ddiger},
  year = {2026},
  eprint = {2606.13194},
  archivePrefix = {arXiv},
  primaryClass = {cs.LG},
  url = {https://arxiv.org/abs/2606.13194}
}

Acknowledgements

This work was partially funded by the IPAI Foundation gGmbH through the Science Residency Program and by the Helmholtz Association Initiative and Networking Fund through the HAICORE@KIT partition. Support was also provided by the HammerHAI project, an EU co-funded AI Factory initiative operated by the High-Performance Computing Center Stuttgart. This project has received funding from the European High Performance Computing Joint Undertaking (EuroHPC JU) under Grant Agreement No. 101234027. It is jointly co-funded by the EuroHPC JU through the European Union's Digital Europe Programme, the European Commission, the German Federal Ministry of Research, Technology and Space (BMFTR), the Baden-Württemberg Ministry of Science, Research and the Arts, the Bavarian State Ministry of Science and the Arts, and the Lower Saxony Ministry of Science and Culture. Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union or the EuroHPC JU.

OpenWearables