SimInsert: 通过区域稀疏注意力融合实现无缝视频对象插入
SimInsert: Seamless Video Object Insertion via Regional Sparse Attention Fusion
AI总结：
提出SimInsert，一种无需训练的视频对象插入方法，利用图像到视频扩散模型的先验和区域稀疏注意力融合，实现时空一致和交互真实感，在PSNR、SSIM和LPIPS上分别提升18.8%、20.1%和降低44.1%。
摘要：
视频对象插入需要确保时空连贯性和交互真实感，远不止简单的内容放置。然而，当前方法通常受限于对显式运动工程或资源密集型重新训练的依赖，限制了其灵活性和泛化能力。为弥补这一差距，我们提出了 extit{SimInsert}，一种无需训练的新范式，将任务高效地分解为直观的单帧编辑和语义运动描述。通过利用图像到视频扩散模型的强大生成先验，SimInsert在时间上传播编辑，严格保持背景不变性，同时实现插入对象与动态环境之间合理的、文本驱动的交互。我们的方法依赖于非侵入式引导机制，这些机制强制执行结构一致性，促进无缝边界融合，并抵消在去噪轨迹中通常累积的保真度漂移。大量定量实验验证了我们的有效性：SimInsert在PSNR上超越最先进方法18.8%，在SSIM上超越20.1%，在LPIPS上降低44.1%，为高保真视频编辑提供了流线型解决方案。
英文摘要：
Video object insertion requires ensuring spatio-temporal coherence and interactive realism, extending far beyond simple content placement. However, current approaches are often hindered by a reliance on explicit motion engineering or resource-intensive retraining, restricting their flexibility and generalization. To bridge this gap, we present \textit{SimInsert}, a training-free paradigm that efficiently decouples the task into intuitive single-frame editing and semantic motion description. By harnessing the robust generative priors of image-to-video diffusion models, SimInsert propagates edits temporally, strictly preserving background invariance while enabling plausible, text-driven interactions between the inserted object and the dynamic environment. Our approach hinges on non-invasive guidance mechanisms that enforce structural consistency, facilitate seamless boundary fusion, and counteract the fidelity drift that typically accumulates during the denoising trajectory. Extensive quantitative experiments validate our efficacy: SimInsert surpasses state-of-the-art methods with an 18.8\% gain in PSNR, 20.1\% in SSIM, and a 44.1\% decrease in LPIPS, offering a streamlined solution for high-fidelity video editing.
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。
