研究人员推出PhyGenHOI，一个旨在生成物理准确且视觉保真的4D人与物体交互（HOI）的新框架。该系统结合了生成式人类运动模型（MDM）和显式物理物体模拟（MPM），使用3D高斯作为统一表示。关键创新包括用于时间同步的窗口吸引损失（Windowed Attraction Loss）、用于动量传递的驱动接触重模拟（Contact-Driven Re-simulation）步骤，以及用于改善接触保真的掩码视频SDS目标（Masked Video-SDS objective）。实验表明，PhyGenHOI能够合成具有逼真交互的动态场景，性能优于现有方法。 AI
影响 该框架有望推动逼真的虚拟环境创建和人机交互模拟。
排序理由 该集群包含一篇详细介绍用于生成4D人与物体交互的新AI框架的研究论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2605.30268v1 Announce Type: cross Abstract: We address the task of generating physically accurate and visually faithful 4D Human-Object Interaction (HOI). Given a static 3D human and target object represented as 3D Gaussian Splats (3DGS), our goal is to synthesize dynamic s…
We address the task of generating physically accurate and visually faithful 4D Human-Object Interaction (HOI). Given a static 3D human and target object represented as 3D Gaussian Splats (3DGS), our goal is to synthesize dynamic scenes where the human actively engages with the ob…
PhyGenHOI synthesizes physically accurate 4D human-object interactions by combining motion diffusion models with material point method simulations using 3D Gaussian representations.