ReconPhys：基于单视频的外观与物理属性重建

摘要

实现物理可信的非刚性物体重建仍是一项重大挑战。现有方法虽能利用可微分渲染进行逐场景优化，恢复几何形态与动态特性，但需耗费大量调参或依赖人工标注，限制了其实用性与泛化能力。为此，我们提出ReconPhys——首个基于单目视频的前馈式框架，可同步学习物理属性估计与3D高斯溅射重建。该方法采用通过自监督策略训练的双分支架构，无需真实物理标签。给定视频序列后，ReconPhys能同步推断几何结构、外观属性与物理参数。在大规模合成数据集上的实验表明其卓越性能：在未来帧预测任务中，本方法以21.64 PSNR显著超越现有优化基线方法的13.27，同时将倒角距离从0.349降低至0.004。关键突破在于，ReconPhys可实现秒级（<1秒）推理，而传统方法需耗时数小时，这为机器人学与图形学领域快速生成可直接仿真的数字资产提供了新途径。

English

Reconstructing non-rigid objects with physical plausibility remains a significant challenge. Existing approaches leverage differentiable rendering for per-scene optimization, recovering geometry and dynamics but requiring expensive tuning or manual annotation, which limits practicality and generalizability. To address this, we propose ReconPhys, the first feedforward framework that jointly learns physical attribute estimation and 3D Gaussian Splatting reconstruction from a single monocular video. Our method employs a dual-branch architecture trained via a self-supervised strategy, eliminating the need for ground-truth physics labels. Given a video sequence, ReconPhys simultaneously infers geometry, appearance, and physical attributes. Experiments on a large-scale synthetic dataset demonstrate superior performance: our method achieves 21.64 PSNR in future prediction compared to 13.27 by state-of-the-art optimization baselines, while reducing Chamfer Distance from 0.349 to 0.004. Crucially, ReconPhys enables fast inference (<1 second) versus hours required by existing methods, facilitating rapid generation of simulation-ready assets for robotics and graphics.

ReconPhys：基于单视频的外观与物理属性重建

ReconPhys: Reconstruct Appearance and Physical Attributes from Single Video

摘要

Support