从单目视频中恢复物理合理的人-物交互
Recovering Physically Plausible Human-Object Interactions from Monocular Videos
把运动学估计的单目人-物交互重建结果送入物理仿真器，用强化学习策略做后处理精修，可复用于任何需要消除穿模与物体漂浮的4D重建管线；其双自更新自适应采样机制也适用于从噪声标注中挑选可靠帧做训练。
原文依据 a physics-guided reconstruction framework that begins with a kinematic estimate and then refines it through a reinforcement learning (RL) policy
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出一种从单目视频中重建物理合理的人-物交互（HOI）的方法。现有基于运动学的方法虽能产生视觉上合理的运动，却常导致相互穿透、物体漂浮等物理伪影。为克服这些问题，我们引入一个物理引导的重建框架：该框架以运动学估计为起点，再通过一个在物理仿真器中训练、用于复现该交互的强化学习（RL）策略对其进行精修。由于运动学估计通常含有噪声，朴素的 RL 训练可能失败。为此，我们提出一种带有双重自更新机制的自适应采样策略，可自动识别出运动学重建信息量最大、最可靠的帧。该过程逐步提升重建质量，并生成物理一致的 HOI 序列。我们在两个标准基准上验证了该方法，在物理合理性指标上相较最优方法取得明显提升。
英文原文摘要 1060 字符
In this paper, we present a method to reconstruct physically plausible human-object interactions (HOI) from monocular videos. While existing kinematic-based approaches produce visually plausible motion, they often result in physical artifacts such as interpenetration and object floating. To overcome these issues, we introduce a physics-guided reconstruction framework that begins with a kinematic estimate and then refines it through a reinforcement learning (RL) policy trained to reproduce the interaction in a physics simulator. Because kinematic estimates are typically noisy, naive RL training can fail. Therefore, we propose an adaptive sampling strategy with a dual self-updating mechanism that automatically identifies the frames with the most informative and reliable kinematic reconstruction. Our process progressively improves reconstruction quality and yields physically consistent HOI sequences. We demonstrate our approach on two standard benchmarks and achieve clear improvements in physical plausibility metrics over state-of-the-art methods.