DEAL-Grasp：用于几何感知灵巧抓取生成的解耦对齐表示

中文摘要 AI 辅助

合成逼真的关节手-物交互是虚拟现实、具身智能和数字人应用中的一个基本问题。现有的灵巧抓取合成方法通常在关节空间中回归或去噪姿态，该空间将全局刚体运动与局部关节运动耦合在一起，这往往会产生不稳定的样本和物理上不合理的接触。我们提出了DEAL-Grasp，它基于解耦对齐（DEAL）表示，将抓取合成重新表述为对齐空间生成：交互状态由任务空间几何锚点和关节参数组成，通过闭式Procrustes对齐恢复刚体变换，同时保留局部关节运动。在此混合状态上，我们使用具有分量式向量场的异构状态流匹配来建模抓取生成，并在训练期间引入时间自适应物理正则化。在推理时，仅通过积分学习到的向量场来合成抓取，无需测试时优化或辅助物理引导。在MultiDex和零样本RealDex基准上，DEAL-Grasp实现了高力扰动成功率，同时具有最小的穿透深度和高度的抓取多样性，并且与依赖优化的基线相比，大幅降低了原生推理延迟。项目页面可在https://wmtlab.github.io/DEAL-Grasp/获取。

英文摘要

Synthesizing realistic articulated hand-object interactions is a fundamental problem in virtual reality, embodied intelligence, and digital human applications. Existing methods for dexterous grasp synthesis typically regress or denoise poses in a joint space that couples global rigid motion with local articulation, which often yields unstable samples and physically implausible contacts. We introduce DEAL-Grasp, built upon the Decoupled Alignment (DEAL) representation, which reformulates grasp synthesis as alignment-space generation: the interaction state comprises task-space geometric anchors and articulation parameters, from which the rigid transform is recovered via closed-form Procrustes alignment while preserving local articulation. On this mixed state, we model grasp generation using heterogeneous-state flow matching with component-wise vector fields, incorporating time-adaptive physical regularization during training. At inference, grasps are synthesized solely by integrating the learned vector field, without test-time optimization or auxiliary physical guidance. Across MultiDex and zero-shot RealDex benchmarks, DEAL-Grasp attains high force-perturbation success rates alongside minimal penetration and high diversity of generated grasps, while substantially reducing native inference latency compared to optimization-heavy baselines. The project page is available at https://wmtlab.github.io/DEAL-Grasp/.

发表机构

• Dalian University of Technology（大连理工大学）

机构由 AI 辅助整理，请以论文原文为准。