SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Choi, Jeongjun, Park, Yeonsoo, Kim, H. Jin
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911368716746752
author Choi, Jeongjun
Park, Yeonsoo
Kim, H. Jin
author_facet Choi, Jeongjun
Park, Yeonsoo
Kim, H. Jin
contents We present SceneNAT, a single-stage masked non-autoregressive Transformer that synthesizes complete 3D indoor scenes from natural language instructions through only a few parallel decoding passes, offering improved performance and efficiency compared to prior state-of-the-art approaches. 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 dedicated triplet predictor for modeling the scene's layout and object relationships by mapping a set of learnable relation queries to a sparse set of symbolic triplets (subject, predicate, object). Extensive experiments on the 3D-FRONT dataset demonstrate that SceneNAT achieves superior performance compared to state-of-the-art autoregressive and diffusion baselines in both semantic compliance and spatial arrangement accuracy, while operating with substantially lower computational cost.
format Preprint
id arxiv_https___arxiv_org_abs_2601_07218
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis
Choi, Jeongjun
Park, Yeonsoo
Kim, H. Jin
Computer Vision and Pattern Recognition
We present SceneNAT, a single-stage masked non-autoregressive Transformer that synthesizes complete 3D indoor scenes from natural language instructions through only a few parallel decoding passes, offering improved performance and efficiency compared to prior state-of-the-art approaches. 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 dedicated triplet predictor for modeling the scene's layout and object relationships by mapping a set of learnable relation queries to a sparse set of symbolic triplets (subject, predicate, object). Extensive experiments on the 3D-FRONT dataset demonstrate that SceneNAT achieves superior performance compared to state-of-the-art autoregressive and diffusion baselines in both semantic compliance and spatial arrangement accuracy, while operating with substantially lower computational cost.
title SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.07218