RigMo：统一绑定与运动学习用于生成式动画
RigMo: Unifying Rig and Motion Learning for Generative Animation
把逐顶点形变编码进「绑定潜变量 + 运动潜变量」两个紧凑潜空间，可前馈地从原始网格序列直接解出高斯骨骼、蒙皮权重与逐帧 SE(3) 变换，为动态 3D 建模提供了一套无需人工骨骼标注、结构可解释的表示方案；在该潜空间上训练 Motion-DiT 即可做下游运动生成，是 4D 生成与角色动画可复用的范式。
原文依据 RigMo encodes per-vertex deformations into two compact latent spaces: a rig latent that decodes into explicit Gaussian bones and skinning weights
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论尽管 4D 生成已取得显著进展，绑定与运动——动画的核心结构与动态组成部分——通常仍被当作彼此独立的问题来建模。现有流程在运动生成时依赖真值骨架与蒙皮权重，并把自动绑定视为一个独立过程，这损害了可扩展性与可解释性。我们提出 RigMo，一个统一的生成式框架，直接从原始网格序列中联合学习绑定与运动，无需任何人工提供的绑定标注。RigMo 将逐顶点形变编码进两个紧凑的潜空间：一个绑定潜变量，可解码为显式的高斯骨骼与蒙皮权重；以及一个运动潜变量，用于产生随时间变化的 SE(3) 变换。二者结合，定义了具有显式结构与连贯运动的可动画网格，从而实现对可形变物体的前馈式绑定与运动推理。除统一的绑定—运动发现之外，我们引入一个在 RigMo 潜空间中运行的 Motion-DiT 模型，并证明这些结构感知的潜变量能够自然地支撑下游运动生成任务。在 DeformingThings4D、Objaverse-XL 与 TrueBones 上的实验表明，RigMo 学习到的绑定平滑、可解释且物理上合理，同时在重建与类别级泛化上优于现有的自动绑定与形变基线。RigMo 为统一、结构感知且可扩展的动态 3D 建模确立了一种新范式。
英文原文摘要 1486 字符
Despite significant progress in 4D generation, rig and motion--the core structural and dynamic components of animation--are typically modeled as separate problems. Existing pipelines rely on ground-truth skeletons and skinning weights for motion generation and treat auto-rigging as an independent process, undermining scalability and interpretability. We present RigMo, a unified generative framework that jointly learns rig and motion directly from raw mesh sequences, without any human-provided rig annotations. RigMo encodes per-vertex deformations into two compact latent spaces: a rig latent that decodes into explicit Gaussian bones and skinning weights, and a motion latent that produces time-varying SE(3) transformations. Together, these outputs define an animatable mesh with explicit structure and coherent motion, enabling feed-forward rig and motion inference for deformable objects. Beyond unified rig-motion discovery, we introduce a Motion-DiT model operating in RigMo's latent space and demonstrate that these structure-aware latents can naturally support downstream motion generation tasks. Experiments on DeformingThings4D, Objaverse-XL, and TrueBones demonstrate that RigMo learns smooth, interpretable, and physically plausible rigs, while achieving superior reconstruction and category-level generalization compared to existing auto-rigging and deformation baselines. RigMo establishes a new paradigm for unified, structure-aware, and scalable dynamic 3D modeling