SceneCritic: 一种用于3D室内场景合成的符号评估器
SceneCritic: A Symbolic Evaluator for 3D Indoor Scene Synthesis
AI总结：
本文提出SceneCritic，一种基于空间本体的符号评估器，用于评估3D室内场景合成的语义、方向和几何一致性，通过不同批评模式验证模型的空间结构构建与修正能力。
摘要：
大型语言模型（LLMs）和视觉-语言模型（VLMs）越来越多地通过布局和场景图等中间结构生成室内场景，但评估仍依赖LLM或VLM裁判，评分易受视角、提示语和幻觉影响。当评估器不稳定时，难以确定模型是否生成了空间合理的场景或输出分数是否反映视角、渲染或提示的选择。我们引入SceneCritic，一种针对平面布局级别的评估器。SceneCritic的约束基于SceneOnto，我们通过聚合3D-FRONT、ScanNet和Visual Genome中的室内场景先验知识构建的结构化空间本体。SceneOnto通过该本体联合验证对象关系中的语义、方向和几何一致性，提供对象级和关系级评估，识别特定违规和成功放置。此外，我们将SceneCritic与迭代细化测试平台结合，探讨模型在不同批评模式下构建和修订空间结构的能力：基于规则的批评使用碰撞约束作为反馈，LLM批评基于布局文本操作，VLM批评基于渲染观测。通过大量实验，我们发现（a）SceneCritic比基于VLM的评估器更符合人类判断，（b）仅文本的LLMs在语义布局质量上可优于VLMs，（c）基于图像的VLM细化是修正语义和方向最有效的批评模式。
英文摘要：
Large Language Models (LLMs) and Vision-Language Models (VLMs) increasingly generate indoor scenes through intermediate structures such as layouts and scene graphs, yet evaluation still relies on LLM or VLM judges that score rendered views, making judgments sensitive to viewpoint, prompt phrasing, and hallucination. When the evaluator is unstable, it becomes difficult to determine whether a model has produced a spatially plausible scene or whether the output score reflects the choice of viewpoint, rendering, or prompt. We introduce SceneCritic, a symbolic evaluator for floor-plan-level layouts. SceneCritic's constraints are grounded in SceneOnto, a structured spatial ontology we construct by aggregating indoor scene priors from 3D-FRONT, ScanNet, and Visual Genome. SceneOnto traverses this ontology to jointly verify semantic, orientation, and geometric coherence across object relationships, providing object-level and relationship-level assessments that identify specific violations and successful placements. Furthermore, we pair SceneCritic with an iterative refinement test bed that probes how models build and revise spatial structure under different critic modalities: a rule-based critic using collision constraints as feedback, an LLM critic operating on the layout as text, and a VLM critic operating on rendered observations. Through extensive experiments, we show that (a) SceneCritic aligns substantially better with human judgments than VLM-based evaluators, (b) text-only LLMs can outperform VLMs on semantic layout quality, and (c) image-based VLM refinement is the most effective critic modality for semantic and orientation correction.
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。
