Human pose estimation (HPE) in privacy-sensitive environments such as healthcare facilities and smart homes demands non-visual sensing solutions. Millimeter-wave (mmWave) radar emerges as a promising alternative, yet its development is hindered by the scarcity of high-fidelity datasets with accurate annotations. This paper introduces mmFree-Pose, the first dedicated mmWave radar dataset specifically designed for privacy-preserving HPE. Collected through a novel visual-free framework that synchronizes mmWave radar with VDSuit-Full motion-capture sensors, our dataset covers 10+ actions, from basic gestures to complex falls. Each sample provides (i) raw 3D point clouds with Doppler velocity and intensity, (ii) precise 23-joint skeletal annotations, and (iii) full-body motion sequences in privacy-critical scenarios. Crucially, all data is captured without the use of visual sensors, ensuring fundamental privacy protection by design. Unlike conventional approaches that rely on RGB or depth cameras, our framework eliminates the risk of visual data leakage while maintaining high annotation fidelity. The dataset also incorporates scenarios involving occlusions, different viewing angles, and multiple subject variations to enhance generalization in real-world applications. By providing a high-quality and privacy-compliant dataset, mmFree-Pose bridges the gap between RF sensing and home monitoring applications, where safeguarding personal identity and behavior remains a critical concern.
We present MambaH-Fit, a state space modelling framework tailored for hyper-surface fitting-based point cloud normal estimation. Existing normal estimation methods often fall short in modelling fine-grained geometric structures, thereby limiting the accuracy of the predicted normals. Recently, state space models (SSMs), particularly Mamba, have demonstrated strong modelling capability by capturing long-range dependencies with linear complexity and inspired adaptations to point cloud processing. However, existing Mamba-based approaches primarily focus on understanding global shape structures, leaving the modelling of local, fine-grained geometric details largely under-explored. To address the issues above, we first introduce an Attention-driven Hierarchical Feature Fusion (AHFF) scheme to adaptively fuse multi-scale point cloud patch features, significantly enhancing geometric context learning in local point cloud neighbourhoods. Building upon this, we further propose Patch-wise State Space Model (PSSM) that models point cloud patches as implicit hyper-surfaces via state dynamics, enabling effective fine-grained geometric understanding for normal prediction. Extensive experiments on benchmark datasets show that our method outperforms existing ones in terms of accuracy, robustness, and flexibility. Ablation studies further validate the contribution of the proposed components.