arXiv：cs.AI· Tzu-Hsin Hsieh, Ricardo Marroquim·· 6 小时前AI 评分38
ElasticFit：基于 VLM 推理与生成式适配的贴合感知 3D 物体插入
ElasticFit: Fit-Aware 3D Object Insertion via VLM Reasoning and Generative Adaptation
AI 导读
ElasticFit 是一个由 VLM 引导的贴合感知 3D 物体插入框架，通过场景接地表示将语言指令转为明确 3D 约束，支持刚性放置、等比缩放和弹性贴合三种适配模式。在固定资产基线对比中，其空间关系成功率从 50.8% 提升至 69.7%，支撑成功率从 48.3% 提升至 91.7%。该工作已被 NeurIPS 2026 接收。
正文
Abstract:Inserting objects into existing 3D scenes requires more than selecting a plausible location:
the inserted object must also fit local geometry while preserving semantic intent and physical plausibility.
Although recent Vision-Language Models (VLMs) and generative models enable semantic reasoning and visual content creation, they offer limited 3D grounding and geometric control when an inserted object must fit into constrained local spaces.
We introduce \textbf{ElasticFit}, a VLM-guided framework for fit-aware object insertion centered on a novel scene-grounded representation.
Given a language instruction and rendered scene observations, ElasticFit infers structured fitting cues that specify where the object should be grounded, what volume it should occupy, how it should be oriented, and its adaptation mode (rigid placement, uniform scaling, or elastic fitting).
These cues convert high-level VLM reasoning into explicit 3D constraints that condition object generation and guide downstream geometric fitting.
ElasticFit then generates a scene-conditioned object prior, reconstructs it in 3D, and refines the mesh through mode-specific fitting while enforcing collision avoidance, contact consistency, and physical grounding.
In fixed-asset baseline comparisons, ElasticFit improves spatial relation success from 50.8\% to 69.7\% and support success from 48.3\% to 91.7\% over the strongest baseline, while providing novel support for generative "make-it-fit" insertions in complex scenarios.
| Comments: | Accepted at NeurIPS 2026. Project page: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.07460 [cs.CV] |
| (or arXiv:2610.07460v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07460 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tzu Hsin Hsieh [view email]
[v1]
Mon, 5 Oct 2026 22:12:18 UTC (47,540 KB)
来源：arXiv：cs.AI · arxiv.org