Saved in:
Bibliographic Details
Main Authors: Bai, Yinuo, Xu, Peijun, Shao, Kuixiang, Jiao, Yuyang, Zhang, Jingxuan, Yao, Kaixin, Gu, Jiayuan, Yu, Jingyi
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.07704
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918379462328320
author Bai, Yinuo
Xu, Peijun
Shao, Kuixiang
Jiao, Yuyang
Zhang, Jingxuan
Yao, Kaixin
Gu, Jiayuan
Yu, Jingyi
author_facet Bai, Yinuo
Xu, Peijun
Shao, Kuixiang
Jiao, Yuyang
Zhang, Jingxuan
Yao, Kaixin
Gu, Jiayuan
Yu, Jingyi
contents Inter-object relations underpin spatial intelligence, yet existing representations -- linguistic prepositions or object-level scene graphs -- are too coarse to specify which regions actually support, contain, or contact one another, leading to ambiguous and physically inconsistent layouts. To address these ambiguities, a part-level formulation is needed; therefore, we introduce PARSE, a framework that explicitly models how object parts interact to determine feasible and spatially grounded scene configurations. PARSE centers on the Part-centric Assembly Graph (PAG), which encodes geometric relations between specific object parts, and a Part-Aware Spatial Configuration Solver that converts these relations into geometric constraints to assemble collision-free, physically valid scenes. Using PARSE, we build PARSE-10K, a dataset of 10,000 3D indoor scenes constructed from real-image layout priors and a curated part-annotated shape database, each with dense contact structures and a part-level contact graph. With this structured, spatially grounded supervision, fine-tuning Qwen3-VL on PARSE-10K yields stronger object-level layout reasoning and more accurate part-level relation understanding; furthermore, leveraging PAGs as structural priors in 3D generation models leads to scenes with substantially improved physical realism and structural complexity. Together, these results show that PARSE significantly advances geometry-grounded spatial reasoning and supports the generation of physically consistent 3D scenes.
format Preprint
id arxiv_https___arxiv_org_abs_2603_07704
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PARSE: Part-Aware Relational Spatial Modeling
Bai, Yinuo
Xu, Peijun
Shao, Kuixiang
Jiao, Yuyang
Zhang, Jingxuan
Yao, Kaixin
Gu, Jiayuan
Yu, Jingyi
Computer Vision and Pattern Recognition
Inter-object relations underpin spatial intelligence, yet existing representations -- linguistic prepositions or object-level scene graphs -- are too coarse to specify which regions actually support, contain, or contact one another, leading to ambiguous and physically inconsistent layouts. To address these ambiguities, a part-level formulation is needed; therefore, we introduce PARSE, a framework that explicitly models how object parts interact to determine feasible and spatially grounded scene configurations. PARSE centers on the Part-centric Assembly Graph (PAG), which encodes geometric relations between specific object parts, and a Part-Aware Spatial Configuration Solver that converts these relations into geometric constraints to assemble collision-free, physically valid scenes. Using PARSE, we build PARSE-10K, a dataset of 10,000 3D indoor scenes constructed from real-image layout priors and a curated part-annotated shape database, each with dense contact structures and a part-level contact graph. With this structured, spatially grounded supervision, fine-tuning Qwen3-VL on PARSE-10K yields stronger object-level layout reasoning and more accurate part-level relation understanding; furthermore, leveraging PAGs as structural priors in 3D generation models leads to scenes with substantially improved physical realism and structural complexity. Together, these results show that PARSE significantly advances geometry-grounded spatial reasoning and supports the generation of physically consistent 3D scenes.
title PARSE: Part-Aware Relational Spatial Modeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.07704