发表机构
Simon Fraser University; Alberta Machine Intelligence Institute (Amii); Canadian Institute for Advanced Research (CIFAR)（西蒙菲莎大学; 阿尔伯塔机器智能研究所; 加拿大高级研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出自监督方法P-CORE，通过确保变形前后基于注意力的点表示的表面一致性，提升其大变形下的鲁棒性，在合成与真实场景编辑任务中均优于现有方法。
AI 中文摘要
神经渲染的进展已实现3D场景的高保真多视图重建，但自由形式的非刚性形状编辑仍是重大挑战。基于点的神经表示因缺乏固定连接性，不会将学习到的表面拓扑结构约束为初始化的拓扑结构，非常适合多视图重建；然而，这一特性导致基于点的表示在大变形下易出现孔洞和表面不连续问题。为解决该问题，本文提出一种新颖的自监督方法，使基于点的表示无需变形几何的真实多视图图像即可适应大变形。核心思路是生成随机变形，确保变形前后预测表面的一致性：具体而言，变形点云的表面预测应与原始点云表面预测经变形后的结果一致。本文将该方法融入基于注意力的点表示中，这类表示与基于高斯核的点表示不同，其使用点间的学习插值核而非每个点周围的高斯核；该学习插值核可学习适应大变形，无需添加或移除点。实验在合成几何编辑基准（Neural Editor、Objaverse）上开展，结果表明本文方法在零样本编辑任务中优于现有基于点的方法，且能显著减少伪影；此外，在DTU和Mip-NeRF 360数据集上的定性结果也验证了本文方法在真实场景中的有效性。
英文摘要
Advances in neural rendering have enabled high-fidelity multi-view reconstruction of 3D scenes. However, free-form non-rigid shape editing remains a significant challenge. Point-based neural representations are highly desirable for multi-view reconstruction because they lack fixed connectivity, which does not constrain the learned surface topology to that of the initialization. Yet this same property causes point-based representations to struggle with holes and surface discontinuities under large deformations. To address this, we propose a novel self-supervised method to enable point-based representations to adapt to large deformations without requiring ground truth multi-view images of deformed geometry. The key idea is to generate random deformations and to ensure consistency in the predicted surface before and after deformation. In particular, the surface prediction from the deformed point cloud should be the same as the deformation applied to the surface prediction from the original point cloud. We incorporate our approach into attention-based point representations, which differ from splatting-based point representations in their use of a learned interpolation kernel between points as opposed to a Gaussian kernel around each point. This learned interpolation kernel can learn to adapt to large deformations, without requiring addition or removal of points. We show that our framework significantly enhances its robustness to large deformations. Experiments on synthetic geometry editing benchmarks (Neural Editor, Objaverse) demonstrate that our approach outperforms existing point-based methods in zero-shot editing and significantly reduces artifacts. Furthermore, qualitative results on the DTU and Mip-NeRF 360 datasets demonstrate our method's effectiveness on real-world scenes.
CommentsAccepted to ECCV 2026. Project Page: https://zvict.github.io/p-core/
