Reproduce “TSFFormer: A Frequency-Domain Spike Transformer for Green Human Activity Recognition” 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
Jiale Deng, Shuiying Xiang
TSFFormer recognizes human activities from wearable inertial measurement unit signals using a spiking model that mixes sensor nodes in the Fourier domain and uses adaptive two-compartment neurons. On WISDM, it reports 86.35% test accuracy, compared with 52.43% for SpikeTCN, the strongest spiking neural network baseline. WISDM uses a specified subject-based 70/15/15 split; results do not establish wearable-device efficiency.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.