EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

Tac-Mamba: A Pose-Guided Cross-Modal State Space Model with Trust-Aware Gating for mmWave Radar Human Activity Recognition

Haiyi Wu, Kai Zhao, Wei Yao, Yong Xiong

Electronics · 2026-04-07 · 15 (7) · 1535

DOI: 10.3390/electronics15071535 ↗

At a glance

Journal article in Electronics, published online on 2026-04-07. Written by Haiyi Wu and 3 co-authors. Cited 2 times so far (Crossref); 32 references. The full text is freely available (open access, CC BY 4.0). Funded by: Shanghai Gerontechnology Innovation Park Research and Development and Testing Public Service Platform Construction Project.

AstrattDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Millimeter-wave (mmWave) radar point clouds offer a privacy-preserving solution for Human Activity Recognition (HAR), but their inherent sparsity and noise limit single-modal performance. While multimodal fusion mitigates this issue, existing methods often suffer from severe negative transfer during visual degradation and incur high computational costs, unsuitable for edge devices. To address these challenges, we propose Tac-Mamba, a lightweight cross-modal state space model. First, we introduce a topology-guided distillation scheme that uses a Spatial Mamba teacher to extract structural priors from visual skeletons. These priors are then explicitly distilled into a Point Transformer v3 (PTv3) radar student with a modality dropout strategy. We also developed a Trust-Aware Cross-Modal Attention (TACMA) module to prevent negative transfer. It evaluates the reliability of visual features through a SiLU-activated cross-modal bilinear interaction, smoothly degrading to a pure radar-driven fallback projection when visual inputs are corrupted. Finally, a Lightweight Temporal Mamba Block (LTMB) with a Zero-Parameter Cross-Gating (ZPCG) mechanism captures long-range kinematic dependencies with linear complexity. Experiments on the public MM-Fi dataset under strict cross-environment protocols demonstrate that Tac-Mamba achieves competitive accuracies of 95.37% (multimodal) and 87.54% (radar-only) with only 0.86M parameters and 1.89 ms inference latency. These results highlight the model’s exceptional robustness to modality missingness and its feasibility for edge deployment.

Funding

  • Shanghai Gerontechnology Innovation Park Research and Development and Testing Public Service Platform Construction Project

Cite

APA

Wu, H., Zhao, K., Yao, W., & Xiong, Y. (2026). Tac-Mamba: A Pose-Guided Cross-Modal State Space Model with Trust-Aware Gating for mmWave Radar Human Activity Recognition. Electronics, 15(7), 1535. https://doi.org/10.3390/electronics15071535

Vancouver

Wu H, Zhao K, Yao W, Xiong Y. Tac-Mamba: A Pose-Guided Cross-Modal State Space Model with Trust-Aware Gating for mmWave Radar Human Activity Recognition. Electronics. 2026;15(7):1535. doi:10.3390/electronics15071535

BibTeX
@article{Wu2026,
  author = {Wu, Haiyi and Zhao, Kai and Yao, Wei and Xiong, Yong},
  title = {Tac-Mamba: A Pose-Guided Cross-Modal State Space Model with Trust-Aware Gating for mmWave Radar Human Activity Recognition},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {7},
  pages = {1535},
  doi = {10.3390/electronics15071535}
}
Start for free →