论文概要
LaxMotion 不使用直接 3D pose 监督，而是把 3D motion 学习为对全局轨迹与单目 2D 运动学线索的一致解释，再以视角一致、朝向连贯和结构稳定的正则进行约束。
研究问题
精确 3D 坐标监督可能鼓励模型拟合训练集中的固定坐标模式，却没有学到跨分布泛化所需的三维结构与动作语义。
论文贡献
- 在 relaxed observability 下重新定义 3D motion generation，不直接回归 3D pose。
- 利用全局轨迹和单目 2D 运动学线索进行结构化 motion factorization。
- 引入 view-consistent alignment、orientation coherence 与 structural stability 的宽松正则。
证据与评测范围
论文报告生成动作具备多样性、时间连贯性和语义对齐，并达到可比或超过全 3D 监督方法的表现；具体数据集、协议和指标应以 v2 实验为准。
适用范围与局限
Relaxed supervision 用轨迹、单目线索和结构正则的假设替代精确坐标目标；在可观测性或相机条件不同的动作领域仍需单独验证。
Related work 定位
LaxMotion 的核心是重新设计 3D 动作生成的监督方式，挑战“更精确的 3D 标签必然带来更好泛化”的默认假设。
Related Work 表述
Liu 等提出 LaxMotion，以全局轨迹、单目运动学线索和宽松结构正则替代直接 3D pose 监督，用于生成三维人体动作。
这是一段用于说明论文定位的简洁中性表述。
论文官方英文摘要
Recent 3D human motion generation models demonstrate remarkable reconstruction accuracy yet struggle to generalize beyond training distributions. This limitation arises partly from the use of precise 3D supervision, which encourages models to fit fixed coordinate patterns instead of learning the essential 3D structure and motion semantic cues required for robust generalization. To overcome this limitation, we propose LaxMotion, a framework that synthesizes realistic 3D motions without direct 3D pose supervision. Instead of regressing toward exact coordinates, LaxMotion learns 3D motion as a consistent explanation of global trajectories and monocular 2D kinematic cues. We introduce a structured motion factorization together with a reformulated training paradigm under relaxed observability. This design is further supported by relaxed regularization objectives that enforce view consistent alignment, orientation coherence, and structural stability. Under this relaxed supervision paradigm, LaxMotion generates diverse, temporally coherent, and semantically aligned 3D motions, achieving performance comparable to or surpassing fully 3D supervised methods. These results indicate that shifting supervision from exact coordinate matching to structural consistency promotes stronger reasoning and improved generalization, offering a scalable and data efficient paradigm for 3D motion generation.
摘要仅用于学术识别，版权仍归论文作者或出版方所有，不属于本页 CC BY 许可范围。
依据与出处
| 核对内容 | 论文中的位置 |
|---|---|
| 问题陈述 | Abstract |
| 方法与贡献 | Abstract; structured motion factorization and relaxed-regularization sections |
| 评测结论 | Abstract; comparisons with fully 3D-supervised methods |
主要核验来源: arXiv record (arXiv:2511.11368v2).
如何引用
科研结论应引用论文本身；只有在复用本站原创解读时才引用本页。
Sheng Liu, Yuanzhi Liang, and Sidan Du. “LaxMotion: Rethinking Supervision Granularity for 3D Human Motion Generation.” European Conference on Computer Vision (ECCV 2026) (2026, forthcoming). arXiv:2511.11368.
复用许可
本页原创解读采用 CC BY 4.0：复用时须署名并链接本页。论文标题、摘要、图表和书目信息不在此许可范围内，仍保留原有权利。 Creative Commons Attribution 4.0 International.