Zero-parameter attention sharing transformer for joint human activity and identity recognition
File(s)
Accepted version
Author(s)
Huang, Shuokang
Chen, Po Yu
Zhou, Peilin
Li, Kaihan
McCann, Julie
Type
Journal Article
Abstract
WiFi-based human sensing is gaining popularity thanks to it not requiring additional devices and it not being as intrusive as cameras. Specifically, human features can be extracted from WiFi Channel State Information (CSI) to recognize human activities, identities, etc. However, most previous works rely on single-task learning models for recognition (e.g., to either recognize activities OR identities solely). The lack of cross-task knowledge sharing restricts these models to task-specific features and poor generalization. Recent studies have applied multi-task learning (MTL) to tackle this, but their cross-task sharing modules add vast amounts of extra parameters. Such massive parameters increase model complexity and reduce time efficiency. In this paper, we propose a novel Zero-parameter Attention Sharing Transformer (ZAST) to efficiently recognize both activities and identities. In ZAST, a Cross-task Attention on Attention (CAoA) mechanism computes the relevance of attention scores for cross-task knowledge sharing, as a new paradigm for lightweight MTL. To mitigate the perturbation caused by attention sharing, we formulate a Multi-head Similarity Loss (L-MS) for stable model training. We further equip ZAST with Channel-wise Squeeze and Excitation (CSE) that efficiently learns the channel correlations of CSI. Extensive experiments on four public datasets indicate that ZAST achieves state-of-the-art recognition performance with the lowest complexity and the highest efficiency.
Date Issued
2025-07-08
Date Acceptance
2025-06-29
Citation
IEEE Transactions on Artificial Intelligence, 2025
ISSN
2691-4581
Publisher
Institute of Electrical and Electronics Engineers
Journal / Book Title
IEEE Transactions on Artificial Intelligence
Copyright Statement
© 2025, IEEE. This is the author’s accepted manuscript made available under a CC-BY licence in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy)
License URL
Publication Status
Published online
Date Publish Online
2025-07-08