PaNDaS：带局部控制的可学习形状插值建模
PaNDaS: Learnable Shape Interpolation Modeling with Localized Control
在源网格上学习逐面特征场并与目标全局编码融合，由形变生成器预测雅可比场再积分出平滑位移，这种「局部特征场 + 雅可比场」的表示可直接迁移到网格/点云的非刚性配准、局部编辑与姿态混合任务，且对重网格化鲁棒。
原文依据 PaNDaS learns a per-face feature field on the source mesh and fuses it with a global encoding of the target.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 PaNDaS，一个用于曲面局部非刚性形变与插值的新型深度学习框架。PaNDaS 在源网格上学习逐面特征场，并将其与目标的全局编码相融合。形变生成器预测一个雅可比场并恢复出平滑位移，从而实现精确的区域控制、姿态混合以及可迁移的局部编辑。与以往方法不同，本方法能够以灵活多样的方式将形变限制在形状的特定部位。在多个身体部位数据集上，PaNDaS 取得了最优的插值精度，并比基于全局形状编码或控制柄的方法具有更强的局部性，同时对重网格化保持鲁棒。我们展示了若干局部形状操控任务，并表明本方法能够通过组合不同的输入形变来生成新的形状。
英文原文摘要 878 字符
We present PaNDaS, a novel deep learning framework for Partial Non-Rigid Deformations and interpolations of Surfaces (PaNDaS). PaNDaS learns a per-face feature field on the source mesh and fuses it with a global encoding of the target. A deformation generator predicts a Jacobian field and recovers a smooth displacement, enabling precise regional control, pose mixing, and transferable local edits. Unlike previous approaches, our method can restrict the deformations to specific parts of the shape in a versatile way. Across various human body part datasets, PaNDaS achieves state-of-the-art interpolation accuracy and stronger locality than methods based on global shape codes or handles, while remaining robust to remeshing. We demonstrate several localized shape manipulation tasks and show that our method can generate new shapes by combining different input deformations.