V-RGBX：对内在属性进行精确控制的视频编辑
V-RGBX: Video Editing with Accurate Controls over Intrinsic Properties
把视频逆渲染得到的内在通道（albedo/normal/material/irradiance）当作可编辑的中间表示，用交错条件机制在关键帧上做物理合理的编辑传播，为可控视频生成提供了「先分解再编辑」的可复用范式。
原文依据 an interleaved conditioning mechanism that enables intuitive, physically grounded video editing through user-selected keyframes
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论大规模视频生成模型在建模真实场景中的逼真外观与光照交互方面已展现出显著潜力。然而，一个能够联合理解场景内在属性（如反照率、法线、材质与辐照度）、利用这些属性进行视频合成、并支持可编辑内在表示的闭环框架仍未被探索。我们提出 V-RGBX，这是首个面向内在属性感知视频编辑的端到端框架。V-RGBX 统一了三项关键能力：（1）将视频逆渲染为内在通道；（2）由这些内在表示合成逼真视频；（3）以内在通道为条件的基于关键帧的视频编辑。V-RGBX 的核心是一种交错条件机制，它通过用户选定的关键帧实现直观且物理合理的视频编辑，并支持对任意内在模态的灵活操控。大量定性与定量结果表明，V-RGBX 能够生成时间一致、逼真的视频，同时以物理合理的方式在序列间传播关键帧编辑。我们展示了其在多种应用中的有效性，包括物体外观编辑与场景级重光照，并超越了先前方法的性能。
英文原文摘要 1297 字符
Large-scale video generation models have shown remarkable potential in modeling photorealistic appearance and lighting interactions in real-world scenes. However, a closed-loop framework that jointly understands intrinsic scene properties (e.g., albedo, normal, material, and irradiance), leverages them for video synthesis, and supports editable intrinsic representations remains unexplored. We present V-RGBX, the first end-to-end framework for intrinsic-aware video editing. V-RGBX unifies three key capabilities: (1) video inverse rendering into intrinsic channels, (2) photorealistic video synthesis from these intrinsic representations, and (3) keyframe-based video editing conditioned on intrinsic channels. At the core of V-RGBX is an interleaved conditioning mechanism that enables intuitive, physically grounded video editing through user-selected keyframes, supporting flexible manipulation of any intrinsic modality. Extensive qualitative and quantitative results show that V-RGBX produces temporally consistent, photorealistic videos while propagating keyframe edits across sequences in a physically plausible manner. We demonstrate its effectiveness in diverse applications, including object appearance editing and scene-level relighting, surpassing the performance of prior methods.