AI 中文总结
针对运动重定向中语义保持和欠定问题，提出ReFM模型，通过SO(3)规范化、对比学习预训练的语义编码器及能量引导的渐进细化流，实现跨角色运动重定向，兼容多种初始化策略。
AI 中文摘要
运动重定向在不同骨骼结构的角色之间传递运动，同时保持语义意图和物理合理性。尽管近期取得了进展，但仍有两个基本问题：(i) 在没有高质量配对重定向数据的情况下，如何学习可靠的源运动语义？(ii) 当没有可靠的配对运动可以作为明确的回归目标时，应如何制定重定向问题？现有方法通常通过将预测约束到复制运动来保持语义。然而，这种初始化将有用的关节线索与由骨骼比例和身体几何不匹配引起的伪影纠缠在一起。此外，在前向传播中直接回归最终运动是受限的，因为重定向本质上是欠定的，期望的解决方案必须平衡语义保真度与目标特定的物理和时间约束，而不是匹配唯一的配对目标。受这些限制的启发，我们提出了ReFM，一种与源网格无关的、能量引导的模型，将运动重定向重新表述为渐进式细化。首先，一个SO(3)规范化器去除冗余的全局方向变化。其次，一个通过对比学习预训练的跨角色语义编码器，为优化指导和语义评估提供角色不变的表示。然后，ReFM通过一个由语义一致性、物理合理性、时间连贯性和最小运动修改引导的学习流，逐步细化初始化的目标运动。该框架兼容不同的初始化策略，包括直接运动复制和Autodesk HumanIK（一种行业标准的全身逆运动学重定向系统）。
英文摘要
Motion retargeting transfers motion across characters with different skeletal structures while preserving semantic intent and physical plausibility. Despite recent progress, two fundamental questions remain: (i) how can reliable source-motion semantics be learned without high-quality paired retargeting data, and (ii) how should retargeting be formulated when no reliable paired motion can serve as a definitive regression objective? Existing methods commonly preserve semantics by constraining predictions toward copied motions. However, such initializations entangle useful articulation cues with artifacts caused by mismatched skeletal proportions and body geometry. Moreover, directly regressing a final motion in one forward pass is restrictive because retargeting is inherently underdetermined, and the desired solution must balance semantic fidelity with target-specific physical and temporal constraints rather than match a unique paired target. Motivated by these limitations, we propose ReFM, a source-mesh-agnostic, energy-guided model that reformulates motion retargeting as progressive refinement. First, an SO(3) canonicalizer removes redundant global-orientation variations. Second, a cross-character semantic encoder, pretrained through contrastive learning, provides a character-invariant representation for both optimization guidance and semantic evaluation. ReFM then progressively refines an initialized target motion through a learned flow guided by semantic consistency, physical plausibility, temporal coherence, and minimal motion modification. The framework is compatible with different initialization strategies, including both direct motion copying and Autodesk HumanIK, an industry-standard full-body inverse-kinematics retargeting system.
