场景画家：基于概念关系对齐的语义一致永续3D场景生成

摘要

永续3D场景生成致力于产出长距离且连贯的3D视角序列，适用于长期视频合成与3D场景重建。现有方法遵循“导航与想象”模式，依赖外推技术实现连续视角扩展。然而，生成视角序列因外推模块累积偏差而遭遇语义漂移问题。为应对这一挑战，我们提出了ScenePainter，一个确保语义一致性的3D场景生成新框架，它将外推器的场景特定先验与当前场景理解对齐。具体而言，我们引入了一种名为SceneConceptGraph的层次图结构，用于构建多层次场景概念间的关系，指导外推器生成一致的新视角，并能动态优化以增强多样性。大量实验证明，我们的框架有效克服了语义漂移问题，生成了更加一致且沉浸感强的3D视角序列。项目页面：https://xiac20.github.io/ScenePainter/。

English

Perpetual 3D scene generation aims to produce long-range and coherent 3D view sequences, which is applicable for long-term video synthesis and 3D scene reconstruction. Existing methods follow a "navigate-and-imagine" fashion and rely on outpainting for successive view expansion. However, the generated view sequences suffer from semantic drift issue derived from the accumulated deviation of the outpainting module. To tackle this challenge, we propose ScenePainter, a new framework for semantically consistent 3D scene generation, which aligns the outpainter's scene-specific prior with the comprehension of the current scene. To be specific, we introduce a hierarchical graph structure dubbed SceneConceptGraph to construct relations among multi-level scene concepts, which directs the outpainter for consistent novel views and can be dynamically refined to enhance diversity. Extensive experiments demonstrate that our framework overcomes the semantic drift issue and generates more consistent and immersive 3D view sequences. Project Page: https://xiac20.github.io/ScenePainter/.

场景画家：基于概念关系对齐的语义一致永续3D场景生成

ScenePainter: Semantically Consistent Perpetual 3D Scene Generation with Concept Relation Alignment

摘要

Support