发表机构
London Digital Twin Research Centre; University of Arkansas; Agora Intelligence Lab（伦敦数字孪生研究中心; 阿肯色大学; 阿戈拉智能实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有WiFi CSI人体姿态预测方法未建模时间动态、误差易累积的问题，提出KOALA框架，通过Koopman潜空间与KAL损失实现无误差累积的多步预测，在MM-Fi和WiPose数据集上性能优于基线。
AI 中文摘要
WiFi信道状态信息（CSI）已成为相机的隐私保护替代方案，可用于人体姿态估计。然而，现有方法将姿态推断视为瞬时回归问题，未对时间动态进行建模，导致无法实现未来运动预测。直接应用基于视觉的预测方法会加剧CSI生成姿态中已存在的估计噪声，因为自回归滚动会在每一步放大误差。我们提出KOALA（一种直接从WiFi CSI预测人体运动的框架），通过将含噪的CSI生成姿态序列提升到学习得到的Koopman潜空间中，使非线性动态变为线性，从而无需自回归迭代或误差累积，仅通过简单的矩阵-向量乘积即可实现多步预测。我们引入了残差CSI条件算子来解决Koopman公式固有的恒定性吸引子问题，并采用锚点-增量预测头消除将当前姿态复制到所有预测步长的退化捷径。为联合正则化提升过程和算子，我们提出了Koopman锚定潜空间（KAL）损失，该损失在时间编码器特征空间中运行，无需对比损失、谱损失或辅助损失即可强制各预测步长间的动态一致性。在MM-Fi和WiPose数据集上的实验表明，KOALA在短期和长期预测步长上均实现了稳健、一致的性能，大幅优于所有基线方法。
英文摘要
WiFi Channel State Information (CSI) has emerged as a privacy-preserving alternative to cameras for human pose estimation. However, existing approaches treat pose inference as an instantaneous regression problem and do not model temporal dynamics, making future motion prediction infeasible. Naively applying vision-based prediction methods compounds the estimation noise already present in CSI-derived poses, as autoregressive rollouts amplify errors at every step. We propose KOALA, the framework for human motion prediction directly from WiFi CSI, by lifting noisy CSI-derived pose sequences into a learned Koopman latent space where nonlinear dynamics become linear, enabling multi-horizon prediction via simple matrix-vector products without autoregressive iteration or error accumulation. A residual CSI-conditioned operator resolves the identity attractor problem inherent from Koopman formulations, and an anchor-delta prediction head eliminates the degenerate shortcut of copying the current pose across all horizons. To regularise the lifting and operator jointly, we introduce a Koopman Anchored Latent (KAL) loss that operates in the temporal-encoder feature space, enforcing dynamical consistency across prediction horizons without requiring contrastive, spectral, or auxiliary losses. Experiments on MM-Fi and WiPose show that KOALA achieves robust, consistent performance across both short- and long-term prediction horizons, outperforming all baselines by a substantial margin.
Comments27 pages, 3 figures
Journal refTransactions on Machine Learning Research (TMLR 2026)
