PoseShield：用于人体自碰撞解析的神经碰撞场

摘要

在基于 SMPL 的人体姿态估计与运动生成中，自碰撞问题始终是持续存在的挑战。在极端关节弯曲或随机运动合成条件下，生成的网格频繁出现自穿透现象，导致物理上不合理的输出结果。我们提出 PoseShield，一种直接在 SMPL 姿态空间中定义的神经碰撞约束。我们将碰撞校正问题形式化为约束优化问题，并将学习到的约束与 Eikonal 方程相关联。通过实施 Eikonal 正则化，确保碰撞边界附近梯度非零，从而提升优化过程的数值稳定性与鲁棒性。与先前在网格空间操作或依赖启发式惩罚项的方法不同，我们的方法直接在人体姿态的低维空间中运行，并具有理论依据。该学习到的约束可进一步扩展至人体运动序列，提供一种与生成模型无关的事后碰撞校正机制，无需重新训练底层运动模型。在一项新构建的 SMPL 姿态基准上的实验表明，我们的方法实现了 95.8% 的成功率，并优于当前最优基线方法。

English

Self-collision remains a persistent challenge in SMPL-based human pose estimation and motion generation. Under extreme articulations or stochastic motion synthesis, generated meshes frequently exhibit self-penetrations, leading to physically implausible results. We propose PoseShield, a neural collision constraint defined directly in SMPL pose space. We formulate collision correction as a constrained optimization problem and connect the learned constraint with the Eikonal equation. Enforcing Eikonal regularization ensures non-vanishing gradients near the collision boundary, improving numerical stability and robustness of the optimization process. Unlike prior methods that operate in the mesh space or rely on heuristic penalties, our approach operates directly in the low-dimensional space of human poses and is theoretically grounded. The same learned constraint extends to human motion sequences, providing a generator-agnostic post-hoc collision corrector without retraining the underlying motion model. Experiments on a newly constructed SMPL pose benchmark show that our method achieves a 95.8% success rate and outperforms state-of-the-art baselines.