Catalyst4D：通过动态传播实现高保真 3D 到 4D 场景编辑
Catalyst4D: High-Fidelity 3D-to-4D Scene Editing via Dynamic Propagation
把静态 3D 编辑结果迁移到动态 4D 高斯场景时，可用「锚点 + 最优传输建立区域级对应」来传播形变，避免跨区域干扰与运动漂移；再用逐高斯颜色不确定性引导选择性外观精修，抑制遮挡导致的时序闪烁。这套锚点对应传播 + 不确定性引导精修的组合可直接复用到任何需要把编辑从规范帧扩散到整段序列的 4D/视频编辑管线。
原文依据 Anchor-based Motion Guidance (AMG) builds a set of structurally stable and spatially representative anchors from both original and edited Gaussians.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期基于神经辐射场与 3DGS 的 3D 场景编辑进展使高质量静态场景编辑成为可能。相比之下，动态场景编辑仍具挑战性，因为直接将 2D 扩散模型扩展到 4D 的方法往往会产生运动伪影、时序闪烁以及不一致的风格传播。我们提出 Catalyst4D，一个将高质量 3D 编辑迁移到动态 4D 高斯场景、同时保持空间与时序一致性的框架。其核心是基于锚点的运动引导（AMG），它从原始高斯与编辑后高斯中构建一组结构稳定且具空间代表性的锚点。这些锚点作为稳健的区域级参考，并通过最优传输建立其对应关系，从而在无跨区域干扰与运动漂移的情况下实现一致的形变传播。作为补充，颜色不确定性引导的外观精修（CUAR）通过估计逐高斯颜色不确定性，并选择性地精修易受遮挡诱发伪影影响的区域，来保持时序外观一致性。大量实验表明，Catalyst4D 实现了时序稳定、高保真的动态场景编辑，并在视觉质量与运动连贯性上优于现有方法。
英文原文摘要 1252 字符
Recent advances in 3D scene editing using NeRF and 3DGS enable high-quality static scene editing. In contrast, dynamic scene editing remains challenging, as methods that directly extend 2D diffusion models to 4D often produce motion artifacts, temporal flickering, and inconsistent style propagation. We introduce Catalyst4D, a framework that transfers high-quality 3D edits to dynamic 4D Gaussian scenes while maintaining spatial and temporal coherence. At its core, Anchor-based Motion Guidance (AMG) builds a set of structurally stable and spatially representative anchors from both original and edited Gaussians. These anchors serve as robust region-level references, and their correspondences are established via optimal transport to enable consistent deformation propagation without cross-region interference or motion drift. Complementarily, Color Uncertainty-guided Appearance Refinement (CUAR) preserves temporal appearance consistency by estimating per-Gaussian color uncertainty and selectively refining regions prone to occlusion-induced artifacts. Extensive experiments demonstrate that Catalyst4D achieves temporally stable, high-fidelity dynamic scene editing and outperforms existing methods in both visual quality and motion coherence.