Reproduce “RF-Net: a Unified Meta-learning Framework for RF-enabled One-shot 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

Shuya Ding, Zhe Chen, Tianyue Zheng, Jun Luo

RF-Net adapts radio-frequency activity recognition to new environments using one labeled example per activity class. It learns across environments, compares support and query observations through weighted feature distances, and adds the base classifier’s logits. The authors report it outperforms tested baselines across Wi-Fi, FMCW radar, and impulse radio, although methods perform similarly on impulse radio; results are from tested settings.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.