VideoCoF：基于时间推理的统一视频编辑
VideoCoF: Unified Video Editing with Temporal Reasoner
AI总结：
VideoCoF通过引入时间推理机制，解决视频编辑中精度与统一性的矛盾，实现无需掩码的精确区域映射和细粒度编辑，验证了方法的高效性与有效性。
摘要：
现有视频编辑方法面临关键权衡：专家模型精度高但依赖任务特定先验如掩码，限制统一性；统一时间上下文学习模型无掩码但缺乏显式空间线索，导致指令到区域映射弱且定位不精确。为解决此冲突，我们提出VideoCoF，一种受链式思维启发的帧链方法。VideoCoF通过强制视频扩散模型先预测推理令牌（编辑区域潜在特征）再生成目标视频令牌，实现“看见、推理、然后编辑”的流程。此显式推理步骤无需用户提供的掩码，同时实现精确的指令到区域对齐和精细视频编辑。此外，我们引入RoPE对齐策略，利用这些推理令牌确保运动对齐并实现训练时长外的长度外推。我们证明，仅需50k视频对的数据成本，VideoCoF在VideoCoF-Bench上达到最先进的性能，验证了我们方法的效率和有效性。我们的代码、权重、数据可在https://github.com/knightyxp/VideoCoF获取。
英文摘要：
Existing video editing methods face a critical trade-off: expert models offer precision but rely on task-specific priors like masks, hindering unification; conversely, unified temporal in-context learning models are mask-free but lack explicit spatial cues, leading to weak instruction-to-region mapping and imprecise localization. To resolve this conflict, we propose VideoCoF, a novel Chain-of-Frames approach inspired by Chain-of-Thought reasoning. VideoCoF enforces a ``see, reason, then edit" procedure by compelling the video diffusion model to first predict reasoning tokens (edit-region latents) before generating the target video tokens. This explicit reasoning step removes the need for user-provided masks while achieving precise instruction-to-region alignment and fine-grained video editing. Furthermore, we introduce a RoPE alignment strategy that leverages these reasoning tokens to ensure motion alignment and enable length extrapolation beyond the training duration. We demonstrate that with a minimal data cost of only 50k video pairs, VideoCoF achieves state-of-the-art performance on VideoCoF-Bench, validating the efficiency and effectiveness of our approach. Our code, weight, data are available at https://github.com/knightyxp/VideoCoF.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
