Abstract
Motion tokenization is a key component of generalizable motion models, yet most existing approaches are restricted to species-specific skeletons, limiting their applicability across diverse morphologies. We propose NECromancer (NEC), a universal motion tokenizer that operates directly on arbitrary BVH skeletons. NEC consists of three components: (1) an Ontology-aware Skeletal Graph Encoder (OwO) that encodes structural priors from BVH files, including joint semantics, rest-pose offsets, and skeletal topology, into skeletal embeddings; (2) a Topology-Agnostic Tokenizer (TAT) that compresses motion sequences into a universal, topology-invariant discrete representation; and (3) the Unified BVH Universe (UvU), a large-scale dataset aggregating BVH motions across heterogeneous skeletons. Experiments show that NEC achieves high-fidelity reconstruction under substantial compression and effectively disentangles motion from skeletal structure. The resulting token space supports cross-species motion transfer, composition, denoising, generation with token-based models, and text-motion retrieval, establishing a unified framework for motion analysis and synthesis across diverse morphologies. Demo page: this https URL
Abstract (translated)
运动标记化是通用运动模型的关键组成部分，然而现有的大多数方法都局限于特定物种的骨骼结构，从而限制了其在不同形态之间的适用性。为此，我们提出了NECromancer（简称NEC），这是一种能在任意BVH骨骼上直接操作的普适运动标记器。NEC由三个部分组成： 1. **Ontology-aware Skeletal Graph Encoder (OwO)**：这一模块根据来自BVH文件的数据编码结构先验知识到骨骼嵌入中，包括关节语义、静止姿态偏移和骨骼拓扑。 2. **Topology-Agnostic Tokenizer (TAT)**：这个组件将运动序列压缩为一种普遍的、与拓扑无关的离散表示形式。 3. **Unified BVH Universe (UvU)**：这是一个大规模的数据集，它汇集了来自不同形态骨骼的各种BVH动作数据。 实验结果表明，NEC在大幅度压缩的情况下仍能实现高保真度的重建，并且有效地将运动与骨骼结构分离。由此产生的标记空间支持跨物种的动作转移、组合、去噪、基于标记模型生成以及文本-动作检索，为不同形态下动作分析和合成提供了一个统一框架。 演示页面: [此链接](this https URL)
URL
https://arxiv.org/abs/2602.06548
