SAM 3D Body：鲁棒的全身人体网格恢复
SAM 3D Body: Robust Full-Body Human Mesh Recovery
提出解耦骨骼姿态与体型的参数化网格表示 MHR，并支持 2D 关键点/掩码提示的编码器-解码器架构，为单图全身（含手、脚）网格恢复提供了可复用的表示与提示式推理范式。
原文依据 It is the first model to use a new parametric mesh representation, Momentum Human Rig (MHR), which decouples skeletal pose and body shape.
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 SAM 3D Body（3DB），一种用于单图全身三维人体网格恢复（HMR）的提示式模型，展现出最优性能，并在多样化的野外条件下具有强泛化性和一致精度。3DB 估计身体、脚和手的姿态。它是首个使用新参数化网格表示 Momentum Human Rig（MHR）的模型，该表示解耦骨骼姿态与体型。3DB 采用编码器-解码器架构，并支持包括 2D 关键点和掩码在内的辅助提示，从而实现类似 SAM 系列模型的用户引导推理。我们从多阶段标注流水线中获取高质量标注，该流水线使用手动关键点标注、可微分优化、多视图几何和稠密关键点检测的各种组合。我们的数据引擎高效地选择和处理数据以确保数据多样性，收集不寻常的姿态和罕见的成像条件。我们提出一个按姿态和外观类别组织的新评测数据集，支持对模型行为进行细致分析。我们的实验在定性用户偏好研究和传统定量分析中均展示出优于先前方法的泛化性和显著改进。3DB 和 MHR 均已开源。
英文原文摘要 1310 字符
We introduce SAM 3D Body (3DB), a promptable model for single-image full-body 3D human mesh recovery (HMR) that demonstrates state-of-the-art performance, with strong generalization and consistent accuracy in diverse in-the-wild conditions. 3DB estimates the human pose of the body, feet, and hands. It is the first model to use a new parametric mesh representation, Momentum Human Rig (MHR), which decouples skeletal pose and body shape. 3DB employs an encoder-decoder architecture and supports auxiliary prompts, including 2D keypoints and masks, enabling user-guided inference similar to the SAM family of models. We derive high-quality annotations from a multi-stage annotation pipeline that uses various combinations of manual keypoint annotation, differentiable optimization, multi-view geometry, and dense keypoint detection. Our data engine efficiently selects and processes data to ensure data diversity, collecting unusual poses and rare imaging conditions. We present a new evaluation dataset organized by pose and appearance categories, enabling nuanced analysis of model behavior. Our experiments demonstrate superior generalization and substantial improvements over prior methods in both qualitative user preference studies and traditional quantitative analysis. Both 3DB and MHR are open-source.