LaRP：基于潜在重投影先验的高效多视角图像修复
LaRP: Efficient Multi-View Inpainting with Latent Reprojection Priors
利用3D基础模型提供的显式跨视角对应关系作为扩散修复模型的条件，实现高效且3D一致的多视角图像修复，可复用于需要多视角一致性的生成任务。
原文依据 we propose a framework that trains an inpainting model to condition on the explicit and reliable multi-view correspondences from a 3D foundation model
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多视角图像修复任务要求修复后的图像具备3D一致性。现有大多数方法先进行单视角2D图像修复，然后在事后3D优化阶段强制多视角一致性，这会导致不理想的伪影和冗长的优化时间。现有的单阶段方法MVInpainter使用视频先验且无需姿态，使其不太适合视频序列之外的输入。本文提出一个框架，训练修复模型以3D基础模型提供的显式且可靠的多视角对应关系为条件。我们框架的核心是跨视角条件化架构LaRP，精心设计以同时利用预训练扩散修复模型的生成先验和重投影的跨视角外观潜在表示。我们还提出一个可扩展的数据管道，用于LaRP的稳定训练。大量实验表明，LaRP在3D一致性和新视角合成质量上优于先前方法，与最先进方法相当，同时速度快50倍。
英文原文摘要 1085 字符
The task of multi-view inpainting necessitates 3D consistency in the inpainted images. Most prior methods first employ single-view 2D inpainting and then enforce multi-view consistency in a post-hoc 3D optimization stage, which leads to undesirable artifacts and lengthy optimization times. The existing single-stage method, MVInpainter, uses video priors and is pose-free, making it less suitable for inputs beyond video sequences. In this paper, we propose a framework that trains an inpainting model to condition on the explicit and reliable multi-view correspondences from a 3D foundation model. Central to our framework is a cross-view conditioning architecture, LaRP, carefully designed to utilize both the generative prior of a pretrained diffusion inpainting model and the reprojected cross-view appearance latents. We additionally propose a scalable data pipeline for stable training of LaRP. Extensive experiments demonstrate that LaRP outperforms prior methods in 3D consistency and novel view synthesis quality competitive with the state-of-the-art, while being 50x faster.