PoseShield: Neural Collision Fields for Human Self-Collision Resolution
2026-06-29 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the problem of self-collision in human pose models, where parts of the body mesh incorrectly overlap during extreme movements. They introduce PoseShield, a method that represents collision avoidance directly in the mathematical space of human poses rather than on the mesh itself. By using a special equation called the Eikonal and solving an optimization problem, their approach improves stability and accuracy of collision correction. Their method works on sequences of motions and does not require retraining existing motion models. Tests show that their approach is more successful than previous methods at preventing self-collisions.
SMPLhuman pose estimationself-collisionpose spaceconstrained optimizationEikonal equationEikonal regularizationmotion generationcollision correctionmesh penetration
Authors
Zhengyuan Li, Zeyun Deng, Yifan Shen, Liangyan Gui, Miaolan Xie, Joseph Campbell, Xifeng Gao, Kui Wu, Zherong Pan, Aniket Bera
Abstract
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.