面向扩散式人体网格恢复的 VLM 引导组偏好对齐
VLM-Guided Group Preference Alignment for Diffusion-based Human Mesh Recovery
把 VLM 驱动的自反思评审智能体产出的细粒度质量分数，转成组间偏好数据用于微调扩散式 HMR，可直接迁移到其他一对多歧义的三维/生成任务（如姿态、深度、新视角合成）做偏好对齐后训练。
原文依据 we propose a group preference alignment framework for finetuning diffusion-based HMR models
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论从单张 RGB 图像恢复人体网格（HMR）本质上是歧义的，因为同一 2D 观测可能对应多个 3D 姿态。近期的扩散式方法通过生成多种假设来应对这一问题，但往往牺牲了精度：其预测要么在物理上不可行，要么偏离输入图像，在遮挡或杂乱的真实场景中尤为明显。为解决这一问题，我们引入了一个具备自反思能力的双记忆增强 HMR 评审智能体，为预测的网格生成上下文感知的质量分数。这些分数提炼了关于 3D 人体运动结构、物理可行性以及与输入图像对齐的细粒度线索。我们利用这些分数构建了一个组级 HMR 偏好数据集。借助该数据集，我们提出了一个组偏好对齐框架，用于微调扩散式 HMR 模型。该过程将丰富的偏好信号注入模型，引导其生成物理上更合理、与图像更一致的人体网格。大量实验表明，与最优方法相比，我们的方法取得了更优的性能。
英文原文摘要 1142 字符
Human mesh recovery (HMR) from a single RGB image is inherently ambiguous, as multiple 3D poses can correspond to the same 2D observation. Recent diffusion-based methods tackle this by generating various hypotheses, but often sacrifice accuracy. They yield predictions that are either physically implausible or drift from the input image, especially under occlusion or in cluttered, in-the-wild scenes. To address this, we introduce a dual-memory augmented HMR critique agent with self-reflection to produce context-aware quality scores for predicted meshes. These scores distill fine-grained cues about 3D human motion structure, physical feasibility, and alignment with the input image. We use these scores to build a group-wise HMR preference dataset. Leveraging this dataset, we propose a group preference alignment framework for finetuning diffusion-based HMR models. This process injects the rich preference signals into the model, guiding it to generate more physically plausible and image-consistent human meshes. Extensive experiments demonstrate that our method achieves superior performance compared to state-of-the-art approaches.