用单步扩散模型实现快速高质量视频编辑，速度提升155倍以上
OSVE: One Step Video Editing with One Step Diffusion Models
- 通过可学习编码器单次前向传播预测帧初始噪声，跳过耗时迭代反演
- 在结构对齐图像对上训练，使编辑保持源视频几何一致性
- 融合跨帧注意力与滑动窗口，保障长视频时间连贯性，适合实时应用
基于文本的视频编辑在扩散模型中效率极低，受限于高成本的多步采样与反演过程。我们提出OSVE，首个成功将单步文生图模型应用于高质量视频编辑的框架，解决了反演、可编辑性和时间一致性三大核心挑战。为避免缓慢的迭代反演，我们训练了一个可学习编码器，在单次前向传播中预测每帧的初始噪声。该编码器在精心构建的结构对齐图像对数据集上，通过新颖的结构感知编辑（SAE）损失进行训练，学习在编辑过程中保持源视频的几何结构。为保证时间连贯性，引入统一帧编辑（UFE）技术，通过拼接帧潜在表示，促进单步生成中的跨帧注意力。对于长视频，采用带锚定帧的滑动窗口策略，维持全局一致性。大量实验表明，OSVE在编辑质量上达到或优于最先进的多步方法，同时运行速度提升约155至171倍。这一突破为实用化的实时视频编辑应用铺平了道路。代码已开源：https://github.com/KU-VGI/OSVE。
原文摘要 · Abstract (English)
Text-guided video editing with diffusion models is impractically slow, hindered by costly multi-step sampling and inversion. We present OSVE, the first framework to successfully adapt one-step Text-to-Image (T2I) models for high-quality video editing, addressing the core challenges of inversion, editability, and temporal consistency. To bypass slow iterative inversion, we train a learnable encoder that predicts the initial noise for each frame in a single forward pass. This encoder is trained with a novel Structure-Aware Editing (SAE) loss on a curated dataset of structurally-aligned image pairs, teaching it to preserve the source video's geometry during edits. For temporal coherence, we introduce Unified-Frame Editing (UFE), a technique that concatenates frame latents to facilitate cross-frame attention in a single generation step. Furthermore, for long videos, a sliding-window strategy with an anchor frame maintains global consistency. Our extensive experiments demonstrate that OSVE achieves editing quality comparable or superior to state-of-the-art multi-step methods, while operating approximately 155--171 times faster. This breakthrough paves the way for practical, real-time video editing applications. Code is available at https://github.com/KU-VGI/OSVE.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。