Unsupervised Kinematic Motion Detection for Part-segmented 3D Shape Collections
Unsupervised Kinematic Motion Detection for Part-segmented 3D Shape Collections
复制标题
DOI：
10.1145/3528233.3530742
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Xianghao Xu;Yifan Ruan;Srinath Sridhar;Daniel Ritchie
中科院分区：
文献类型：
--
作者：
Xianghao Xu;Yifan Ruan;Srinath Sridhar;Daniel Ritchie
制造对象的3D模型对于填充虚拟世界以及视觉和机器人技术的合成数据生成非常重要。最有用的是，这些对象应该是铰接的：当与它们交互时，它们的部分应该移动。虽然存在铰接对象数据集，但创建它们是劳动密集型的。基于学习的零件运动预测可以提供帮助，但所有现有方法都需要带注释的训练数据。在本文中，我们提出了一种无监督的方法来发现关节运动的部分分割的3D形状集合。我们的方法是基于一个概念，我们称之为类别封闭：任何有效的表达一个对象的部分应该保持在同一个语义类别的对象（例如，椅子保持椅子）。我们操作这个概念的算法，优化形状的部分运动参数，使它可以转换成其他形状的集合。我们评估我们的方法，使用它来重新发现部分运动的PartNet-Mobility数据集。对于几乎所有的形状类别，我们的方法的预测运动参数相对于地面实况注释具有较低的误差，优于两种监督运动预测方法。
3D models of manufactured objects are important for populating virtual worlds and for synthetic data generation for vision and robotics. To be most useful, such objects should be articulated: their parts should move when interacted with. While articulated object datasets exist, creating them is labor-intensive. Learning-based prediction of part motions can help, but all existing methods require annotated training data. In this paper, we present an unsupervised approach for discovering articulated motions in a part-segmented 3D shape collection. Our approach is based on a concept we call category closure: any valid articulation of an object’s parts should keep the object in the same semantic category (e.g. a chair stays a chair). We operationalize this concept with an algorithm that optimizes a shape’s part motion parameters such that it can transform into other shapes in the collection. We evaluate our approach by using it to re-discover part motions from the PartNet-Mobility dataset. For almost all shape categories, our method’s predicted motion parameters have low error with respect to ground truth annotations, outperforming two supervised motion prediction methods.