NeuROK：生成式 4D 神经物体运动学
NeuROK: Generative 4D Neural Object Kinematics
将物体动力学参数化为低维潜空间，并用 Transformer 编码器-解码器从潜变量解码变形形状，为 4D 动态生成提供可复用的数据驱动框架，避免依赖预定义物理模型。
原文依据 we learn both a latent space of all possible states of the object and a decoder that maps any sampled latent to a plausibly deformed shape
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论数据驱动方法已经彻底改变了 3D 视觉，使 Transformer 能够有效地重建和生成静态 3D 物体。然而，生成模拟的 4D 动力学——即静态物体在各种物理条件下逼真的时序变形——仍然具有挑战性且常常是临时的，尽管这对于构建全面的 3D 世界模型至关重要。大多数现有方法假设一个预定义的物理模型，并使用系统辨识来估计参数，从而将这些方法限制在特定类别和小规模数据集上。我们提出，这些限制可以通过学习以物体为中心的物理系统的数据驱动运动学状态参数化来克服。具体来说，我们学习物体所有可能状态的潜空间，以及一个解码器，该解码器将任何采样的潜变量映射到物体合理的变形形状。我们将这种参数化称为神经物体运动学（NeuROK），并在一个精心整理的大规模 4D 数据集上学习基于 Transformer 的编码器-解码器模型。这种公式化和学习到的模型显著简化了模拟动力学的生成，因为从经典物理中的拉格朗日力学视角来看，我们只需要考虑低维潜空间内的动力学。我们在多种动态物体类型上展示了该框架的有效性和通用性，显示出与先前工作相比的明显优势。
英文原文摘要 1424 字符
Data-driven approaches have revolutionized 3D vision, enabling transformers to effectively reconstruct and generate static 3D objects. However, generating simulative 4D dynamics---realistic temporal deformations of static objects under various physical conditions---remains challenging and often ad hoc despite being critical for building comprehensive 3D world models. Most existing methods assume a predefined physical model and use system identification to estimate parameters, restricting these methods to specific categories and small-scale datasets. We propose that these restrictions can be overcome by learning a data-driven kinematic state parameterization for object-centric physical systems. Specifically, we learn both a latent space of all possible states of the object and a decoder that maps any sampled latent to a plausibly deformed shape of the object. We refer to this parameterization as Neural Object Kinematics (NeuROK), and learn a transformer-based encoder-decoder model on a curated large-scale 4D dataset. This formulation and the learned model significantly simplify the generation of simulative dynamics since we only need to consider the dynamics within a low-dimensional latent space from the Lagrangian mechanics' perspective in classical physics. We demonstrate the effectiveness and generality of this framework across diverse dynamic object types, showing clear advantages over prior works.