MAMMA：无标记高精度多人动作捕捉
MAMMA: Markerless Accurate Multi-person Motion Acquisition
以分割掩码为条件、用可学习 query 预测稠密 2D 接触/可见性感知表面关键点，可在严重遮挡与人体交互下建立逐人对应，为多视角 SMPL-X 拟合提供鲁棒 2D 监督与对应关系。
原文依据 predicts dense 2D contact-aware and visibility-aware surface landmarks conditioned on segmentation masks
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 MAMMA，一种无标记动作捕捉流程，可从多视角视频中准确恢复 SMPL-X 参数。传统动作捕捉系统依赖物理标记。尽管其精度很高，但对专用硬件、人工粘贴标记以及大量后处理的需求使其成本高昂且耗时。近期的基于学习的方法试图克服这些限制，但大多面向单人捕捉、依赖稀疏关键点，或在遮挡与人体物理交互下表现不佳。在本工作中，我们引入一种方法：以分割掩码为条件，预测稠密的、接触感知且可见性感知的 2D 表面关键点，从而即使在严重遮挡下也能实现逐人的对应估计。我们采用一种新颖架构，为每个关键点利用可学习 query。我们证明该方法能够处理复杂的人与人交互，并比现有方法具有更高精度。为训练网络，我们构建了一个大规模合成多视角数据集，融合来自多种来源的人体动作，包括极端姿态、手部动作和近距离交互。该数据集产生高变异性的合成序列，具有丰富的身体接触与遮挡，并包含带有稠密 2D 关键点标注的 SMPL-X 真值。最终得到的系统无需标记即可准确捕捉人体动作。与商业化的基于标记的动作捕捉方案相比，我们的方法在无需大量人工清理的情况下提供了有竞争力的重建质量。最后，针对稠密关键点预测与无标记动作捕捉缺乏通用基准的问题，我们基于真实多视角序列引入了两种评测设定。
英文原文摘要 1807 字符
We present MAMMA, a markerless motion-capture pipeline that accurately recovers SMPL-X parameters from multi-view video. Traditional motion-capture systems rely on physical markers. Although they offer high accuracy, their requirements of specialized hardware, manual marker placement, and extensive post-processing make them costly and time-consuming. Recent learning-based methods attempt to overcome these limitations, but most are designed for single-person capture, rely on sparse keypoints, or struggle with occlusions and physical interactions. In this work, we introduce a method that predicts dense 2D contact-aware and visibility-aware surface landmarks conditioned on segmentation masks, enabling person-specific correspondence estimation even under heavy occlusion. We employ a novel architecture that exploits learnable queries for each landmark. We demonstrate that our approach can handle complex person--person interaction and offers greater accuracy than existing methods. To train our network, we construct a large, synthetic multi-view dataset combining human motions from diverse sources, including extreme poses, hand motions, and close interactions. Our dataset yields high-variability synthetic sequences with rich body contact and occlusion, and includes SMPL-X ground-truth annotations with dense 2D landmarks. The result is a system capable of accurately capturing human motion without the need for markers. Our approach offers competitive reconstruction quality compared to commercial marker-based motion-capture solutions, without the extensive manual cleanup. Finally, we address the absence of common benchmarks for dense-landmark prediction and markerless motion capture by introducing two evaluation settings built from real multi-view sequences. https://mamma.is.tue.mpg.de/