arXiv:2607. 26381v1 Announce Type: cross Abstract: Wi-Fi-based human activity recognition has advanced substantially, but most existing methods assume a closed set of activities and require labeled Wi-Fi samples for every target class, limiting their ability to recognize unseen activities.
By Yitong Shen, Cheng Guo, Peiliang Wang, Jingzhe Zhang, Yi Sheng, Haopeng Zhang, Hongfei Xue, Yili Ren
arXiv:2606. 01834v1 Announce Type: cross Abstract: Human Action Recognition (HAR) using WiFi Channel State Information (CSI) has gained increasing attention due to its non-contact, low-cost, and privacy-preserving nature.
By Chinthaka Ranasingha, Tharindu Fernando, Sridha Sridharan, Clinton Fookes, Harshala Gammulle
CIG-MAE is a self‑supervised framework for WiFi‑based human action recognition that uses a cross‑modal masked autoencoder to reconstruct both amplitude and phase of Channel State Information. It introduces an adaptive, information‑guided masking strategy that focuses on high‑density time‑frequency regions and employs a Barlow Twins regularizer to align cross‑modal representations without negative samples. Experiments on three public datasets show that CIG‑MAE outperforms state‑of‑the‑art SSL methods and even surpasses a fully supervised baseline, highlighting its data efficiency, robustness, and generalization.
By Gang Liu, Yanling Hao, Yixuan Zou
arXiv:2608.08381v2 Announce Type: replace
Abstract: Motivated by the IEEE 802.11bf effort to standardize advanced WLAN sensing, interest in Wi-Fi Channel State Information (CSI) for passive, device-f...
By Navid Hasanzadeh, Shahrokh Valaee
Zero-shot learning (ZSL) for inertial measurement unit (IMU)-based human activity recognition (HAR) faces a central challenge: bridging the gap between sensor embeddings and semantic class representations. We systematically evaluate seven configurations combining three inference methods with two training pipelines on the PAMAP2 dataset, using 14 seen and 4 unseen activity classes with subjects 108 and 109 held out for testing.
BayaHAR is a lightweight, gradient‑free framework that adapts pretrained sensor‑based Human Activity Recognition (HAR) classifiers to new users by converting them into Prototypical Networks with prior prototypes that maintain zero‑shot performance. It introduces closed‑form Bayesian prototype estimation for labeled calibration data and extends this approach to weakly labeled data, requiring only activity labels. With just three seconds of calibration per activity, supervised adaptation boosts test macro‑F1 on unseen users by 2.76–33.44 percentage points across four datasets, while weakly supervised adaptation improves by 0.56–32.13 points, enabling efficient on‑device personalization.
By Maximilian Burzer, Till Riedel, Michael Beigl, Tobias R\"oddiger