用自然语言快速生成高质量3D室内场景，效率远超此前方法。
SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis
- 采用掩码非自回归架构，一次并行生成完整场景。
- 在3D-FRONT数据集上，语义契合度与空间布局准确率均领先。
- 适合需要大规模生成多样化3D场景的应用场景。
我们提出SceneNAT，一种基于自然语言指令的3D室内场景生成模型。该模型采用掩码非自回归Transformer，在少量并行解码步骤内完成场景生成，显著提升生成质量与效率。通过在语义和空间属性的完全离散表示上进行掩码建模，并在属性级与实例级同时应用掩码策略，模型能更好捕捉物体内与物体间结构。为增强关系推理，引入关系推理模块（RRM），将关系建模转化为集合预测任务，提取结构感知特征以引导布局生成，无需显式序列解析。在3D-FRONT数据集上的实验表明，SceneNAT在语义一致性与空间排列准确性方面优于当前最优的自回归与扩散基线方法，且计算开销大幅降低，可实现大规模、高吞吐量的多样化场景生成。
原文摘要 · Abstract (English)
We present SceneNAT, a masked non-autoregressive Transformer for 3D indoor scene synthesis from natural language instructions. It generates complete scenes in a few parallel decoding passes, improving both quality and efficiency over prior methods. SceneNAT is trained via masked modeling over fully discretized representations of both semantic and spatial attributes. By applying a masking strategy at both the attribute level and the instance level, the model can better capture intra-object and inter-object structure. To boost relational reasoning, SceneNAT employs a relational reasoning module (RRM) that captures implicit spatial constraints. By formulating relation modeling as a set prediction task, it extracts structure-aware features to guide the layout generation without explicit sequential parsing. Extensive experiments on 3D-FRONT show that SceneNAT outperforms state-of-the-art autoregressive and diffusion baselines in both semantic compliance and spatial arrangement accuracy while using substantially lower computational cost, enabling high-throughput generation of diverse scenes at scale.
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。