基于高斯溅射构建复杂铰接物体的可交互复制体

Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting

February 26, 2025

作者: Yu Liu, Baoxiong Jia, Ruijie Lu, Junfeng Ni, Song-Chun Zhu, Siyuan Huang

cs.AI

摘要

构建关节物体是计算机视觉领域的一项关键挑战。现有方法往往难以有效整合不同物体状态间的信息，限制了部件网格重建和部件动态建模的准确性，尤其对于复杂的多部件关节物体而言。我们提出了ArtGS，一种新颖的方法，利用3D高斯作为灵活且高效的表示来解决这些问题。我们的方法结合了规范高斯与从粗到细的初始化和更新策略，以对齐不同物体状态下的关节部件信息，并采用了一种受蒙皮启发的部件动态建模模块，以提升部件网格重建和关节学习的效果。在合成和真实世界数据集上的大量实验，包括针对复杂多部件物体的新基准测试，均表明ArtGS在联合参数估计和部件网格重建方面达到了最先进的性能。我们的方法显著提高了重建质量和效率，特别是对于多部件关节物体。此外，我们还提供了对设计选择的全面分析，验证了每个组件的有效性，并指出了未来改进的潜在方向。

English

Building articulated objects is a key challenge in computer vision. Existing methods often fail to effectively integrate information across different object states, limiting the accuracy of part-mesh reconstruction and part dynamics modeling, particularly for complex multi-part articulated objects. We introduce ArtGS, a novel approach that leverages 3D Gaussians as a flexible and efficient representation to address these issues. Our method incorporates canonical Gaussians with coarse-to-fine initialization and updates for aligning articulated part information across different object states, and employs a skinning-inspired part dynamics modeling module to improve both part-mesh reconstruction and articulation learning. Extensive experiments on both synthetic and real-world datasets, including a new benchmark for complex multi-part objects, demonstrate that ArtGS achieves state-of-the-art performance in joint parameter estimation and part mesh reconstruction. Our approach significantly improves reconstruction quality and efficiency, especially for multi-part articulated objects. Additionally, we provide comprehensive analyses of our design choices, validating the effectiveness of each component to highlight potential areas for future improvement.