发表机构
The Hong Kong Polytechnic University; Tencent ARC Lab（香港理工大学; 腾讯ARC实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
OREO提出一种动态优化框架，利用二维扩散先验和强化编辑，以即时渲染编辑作为伪目标，提升三维生成模型的视觉保真度。
AI 中文摘要
尽管三维生成领域近期取得了进展，但模型往往难以生成具有高视觉保真度的资产。为弥合这一差距，我们提出了OREO，一种通过对齐框架，利用丰富的二维扩散先验来增强三维生成器的真实感。OREO不依赖静态数据集，而是建立了一个动态优化循环，将即时编辑的渲染结果作为二维伪目标。其核心在于我们引入了强化编辑（Reinforced Editing），该方法利用二维模型来细化三维输出的渲染视图，在保留底层几何、视角和内容的同时，提升其整体视觉保真度。这些细化后的视图作为高质量监督目标，使三维生成器能够从自身生成的样本中学习，并逐步提升其视觉质量。实验表明，OREO能有效改进预训练基线模型，生成具有增强视觉真实感的三维资产。
英文摘要
Despite recent advancements in 3D generation, models often struggle to produce assets with high visual fidelity. To bridge this gap, we propose OREO, an alignment framework that enhances the realism of 3D generators by leveraging rich 2D diffusion priors. Instead of relying on static datasets, OREO establishes a dynamic optimization loop that produces on-the-fly edited renderings as 2D pseudo-targets. At its core, we introduce Reinforced Editing, which utilizes a 2D model to refine rendered views of the 3D output, enhancing their overall visual fidelity while preserving the underlying geometry, viewpoint, and content. These refined views serve as high-quality supervision targets, enabling the 3D generator to learn from its own generated samples and progressively improve its visual quality. Experiments demonstrate that OREO effectively improves upon pre-trained baselines, producing 3D assets with enhanced visual realism.
CommentsAccepted to ECCV 2026. Our project page is at https://theericma.github.io/oreo/
