提升3D布局生成中相对空间关系的可靠性与一致性
R$^3$L: Reasoning 3D Layouts from Relative Spatial Relations
- 通过不变空间分解打破耦合关系链，减少误差累积
- 采用想象-修正循环实现自洽的空间推理，提升布局合理性
- 适合需要精准空间逻辑的3D场景生成任务
相对空间关系为三维布局结构提供了紧凑表示，是相对空间推理的基础。现有方法利用多模态大语言模型推断此类关系，但推断结果常不可靠，通常依赖事后启发式处理。本文提出R$^3$L框架，提升3D布局生成中相对空间推理的可靠性和一致性。核心动机是：多跳推理需反复进行参考系变换，导致关系推断误差累积，引发语义与度量漂移。为此，我们提出不变空间分解以解耦关系链，并引入一致空间想象机制，通过想象-修正循环增强自洽性。此外，设计支持性空间优化，通过全局到局部坐标重参数化缓解位姿优化难度。在多种场景类型和指令下进行的大量实验表明，R$^3$L生成的布局更符合物理现实且语义一致。分析显示，解决帧引起的不一致性对可靠多跳推理至关重要。代码已开源。
原文摘要 · Abstract (English)
Relative spatial relations provide a compact representation of spatial structure and are fundamental to relative spatial reasoning in 3D layout generation. Recent works leverage Multimodal Large Language Models (MLLMs) to infer such relations, but the inferred relations are often unreliable and are typically handled with post-hoc heuristics. In this paper, we propose R$^3$L, a general framework that improves the reliability and consistency of relative spatial reasoning for 3D layout generation. Our key motivation is that multi-hop reasoning requires repeated reference-frame transformations, which accumulate errors in inferred relations and lead to semantic and metric drift. To mitigate this, we propose invariant spatial decomposition to break coupled relation chains, and consistent spatial imagination to promote self-consistency through an imagine-and-revise loop. We further introduce supportive spatial optimization to ease pose optimization via global-to-local coordinate re-parameterization. Extensive experiments across diverse scene types and instructions demonstrate that R$^3$L produces more physically feasible and semantically consistent layouts. Notably, our analysis shows that resolving frame-induced inconsistencies is crucial for reliable multi-hop relative spatial reasoning. The code is available at https://github.com/Neal2020GitHub/R3L.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。