arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.26241cs.CV
收藏

RoMo:用于人体运动生成的大规模、丰富组织的数据集和语义分类体系

RoMo: A Large-Scale, Richly Organized Dataset and Semantic Taxonomy for Human Motion Generation

Australian National University(澳大利亚国立大学);Roblox;Stanford University(斯坦福大学);Rutgers University(罗格斯大学)

Jiahao Zhang, Joseph Liu, Young-Yoon Lee, Seonghyeon Moon, Victor Zordan, Guy Tevet, Karen Liu, Stephen Gould, Oren Jacob, Haomiao Jiang, Mubbasir Kapadia, Yizhak Ben-Shabat

AI总结:

提出RoMo数据集,通过分类感知过滤流水线确保质量,并采用三级语义分类体系组织数据,使训练模型在保真度和多样性上达到最优,同时提升对复杂文本提示的理解。

摘要:

在语言、图像和视频领域的生成建模成功表明,大型、精心策划的数据集是构建强大模型的关键驱动力。然而,3D人体运动领域一直滞后,受限于在小型高保真运动捕捉数据集和以静态或低质量序列为主的大规模野外数据集之间的不满意选择。我们引入了RoMo,一个丰富、大规模、精心策划的野外人体运动数据集,解决了这些权衡。为确保质量,我们引入了一个分类感知过滤流水线,积极去除静态和易产生伪影的序列。每个序列都带有详细注释,并由一个新颖的三级语义分类体系组织。这种层次结构实现了细粒度的逐类别评估,揭示了全局指标所掩盖的模型优势和弱点。我们证明,在RoMo上训练的模型在保真度和多样性上达到最优,同时获得了对复杂、细微文本提示的卓越理解。最后,我们发布了Motion Toolbox以标准化指标、数据转换和可视化,为可重复和可解释的运动生成研究奠定了基础。

英文摘要:

Success in generative modeling across language, image, and video demonstrates that large, well-curated datasets are the key driver for building capable models. 3D Human motion, however, has lagged behind, constrained by an unsatisfying choice between small, high-fidelity motion capture datasets and large-scale in-the-wild collections dominated by static or low-quality sequences. We introduce RoMo, a rich, large-scale, carefully curated dataset of in-the-wild human motions that resolves these tradeoffs. To ensure quality, we introduce a taxonomy-aware filtering pipeline that aggressively removes static and artifact-prone sequences. Every sequence is annotated with detailed captions and organized by a novel three-level semantic taxonomy. This hierarchical structure enables fine-grained, per-category evaluation, that reveals model strengths and weaknesses obscured by global metrics. We demonstrate that models trained on RoMo achieve state-of-the-art fidelity and diversity while gaining a superior understanding of complex, subtle text prompts. Finally, we release the Motion Toolbox to standardize metrics, data conversion, and visualization, establishing a foundation for reproducible and interpretable motion generation research.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑