动作基元：人体运动的自监督分层表示
Action Motifs: Self-Supervised Hierarchical Representation of Human Body Movements
把 3D 人体姿态序列切成变长片段、每段压成一个潜变量词元，再用掩码词元预测自监督地学出可复用的动作基元（Action Motifs），这种分层潜变量 Transformer 可直接迁移到动作识别、运动预测与运动插值的表征预训练中。
原文依据 A4Mer splits a 3D pose sequence into variable-length segments and represents each segment as a single latent token (Action Atoms).
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论有效的人类行为建模需要一种能够利用其组合性的人体运动表示。我们提出一种分层表示，由捕捉原子关节运动的动作原子（Action Atoms）和由这些原子经时序组合而成、编码在不同整体人类动作中出现的相似身体运动的动作基元（Action Motifs）构成。我们由此推导出 A4Mer，一种嵌套潜变量 Transformer，以完全自监督的方式从人体姿态数据中学习该分层表示。A4Mer 将 3D 姿态序列切分为变长片段，并将每个片段表示为单个潜变量词元（动作原子）。通过自底向上的表示学习，由这些动作原子构成的时序模式自然涌现（动作基元），它们捕捉了可复用的、语义化的身体运动片段中有意义的时序跨度。A4Mer 通过在各自潜变量空间中统一的掩码词元预测 pretext 任务实现这一点。我们还引入了动作基元数据集（Action Motif Dataset, AMD），这是一个带有完整 SMPL 标注的大规模多视角人类行为视频数据集。我们提出了一种新颖的相机使用方式，将相机安装在脚上，从而在频繁且严重的身体遮挡情况下仍能获得逐帧标注。实验结果表明，A4Mer 能有效提取有意义的动作基元，并显著有益于人类行为建模任务，包括动作识别、运动预测和运动插值。
英文原文摘要 1497 字符
Effective human behavior modeling requires a representation of the human body movement that capitalizes on its compositionality. We propose a hierarchical representation consisting of Action Atoms that capture the atomic joint movements and Action Motifs that are formed by their temporal compositions and encode similar body movements found across different overall human actions. We derive A4Mer, a nested latent Transformer to learn this hierarchical representation from human pose data in a fully self-supervised manner. A4Mer splits a 3D pose sequence into variable-length segments and represents each segment as a single latent token (Action Atoms). Through bottom-up representation learning, temporal patterns composed of these Action Atoms, which capture meaningful temporal spans of reusable, semantic segments of body movements, naturally emerge (Action Motifs). A4Mer achieves this with a unified pretext task of masked token prediction in their respective latent spaces. We also introduce Action Motif Dataset (AMD), a large-scale dataset of multi-view human behavior videos with full SMPL annotations. We introduce a novel use of cameras by mounting them on the feet to achieve their frame-wise annotations despite frequent and heavy body occlusions. Experimental results demonstrate the effectiveness of A4Mer for extracting meaningful Action Motifs, which significantly benefit human behavior modeling tasks including action recognition, motion prediction, and motion interpolation.