Reproduce “Deep Learning for Sensor-based Activity Recognition: A Survey” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jindong Wang, Yiqiang Chen, Shuji Hao, Xiaohui Peng, Lisha Hu
The authors survey deep-learning work in sensor-based human activity recognition, organizing it by sensor modality, model, and application. They explain how networks learn representations from readings rather than relying on hand-designed features, then synthesize studies. On OPPORTUNITY protocol 1, Table 5 reports weighted F1 scores of 92.70 for b-LSTM-S and 88.30 for CNN; scores vary by setting, and no model wins everywhere.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.