SceneCritic: A Symbolic Evaluator for 3D Indoor Scene Synthesis

• URL: http://arxiv.org/abs/2604.13035v1
• Date: Tue, 14 Apr 2026 17:59:26 GMT
• Title: SceneCritic: A Symbolic Evaluator for 3D Indoor Scene Synthesis
• Abstract summary: We introduce SceneCritic, a symbolic evaluator for floor-plan-level layouts.<n> SceneCritic's constraints are grounded in SceneOnto, a structured spatial ontology.<n>We show that SceneCritic aligns substantially better with human judgments than VLM-based evaluators.
• Score: 6.182838033073788
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) and Vision-Language Models (VLMs) increasingly generate indoor scenes through intermediate structures such as layouts and scene graphs, yet evaluation still relies on LLM or VLM judges that score rendered views, making judgments sensitive to viewpoint, prompt phrasing, and hallucination. When the evaluator is unstable, it becomes difficult to determine whether a model has produced a spatially plausible scene or whether the output score reflects the choice of viewpoint, rendering, or prompt. We introduce SceneCritic, a symbolic evaluator for floor-plan-level layouts. SceneCritic's constraints are grounded in SceneOnto, a structured spatial ontology we construct by aggregating indoor scene priors from 3D-FRONT, ScanNet, and Visual Genome. SceneOnto traverses this ontology to jointly verify semantic, orientation, and geometric coherence across object relationships, providing object-level and relationship-level assessments that identify specific violations and successful placements. Furthermore, we pair SceneCritic with an iterative refinement test bed that probes how models build and revise spatial structure under different critic modalities: a rule-based critic using collision constraints as feedback, an LLM critic operating on the layout as text, and a VLM critic operating on rendered observations. Through extensive experiments, we show that (a) SceneCritic aligns substantially better with human judgments than VLM-based evaluators, (b) text-only LLMs can outperform VLMs on semantic layout quality, and (c) image-based VLM refinement is the most effective critic modality for semantic and orientation correction.

Related papers

• Human-Centric Image Captioning with Subject-Centered Spatial Understanding [27.382505649043637]
  We introduce SPACE, a benchmark designed to evaluate subject-centered spatial understanding.<n>We propose a specialized data construction and alignment pipeline to bridge the gap.<n>Our framework significantly improves human-centric caption quality, particularly in subject-centered spatial reasoning.
  arXiv  Detail & Related papers  (2026-09-08T06:21:32Z)
• IDEAL-Bench: Indoor Dataset and Evaluation suite for Analyzing 3D Layout reasoning [30.654858163397662]
  We introduce I-Bench, an evaluation suite that requires Vision-Language Models to predict structured 3D layouts on indoor scenes.<n>I-Bench moves beyond CLEVR-style simple geometric primitives so that any image-space discrepancy reflects spatial reasoning alone.
  arXiv  Detail & Related papers  (2026-07-03T22:04:42Z)
• Compact Object-Level Representations with Open-Vocabulary Understanding for Indoor Visual Relocalization [93.79141221821858]
  OpenReLoc is a camera relocalization system designed to provide scene understanding and accurate pose estimation capabilities.<n>We propose a multi-modal mechanism to integrate open-vocabulary semantic knowledge for effective 2D-3D object matching.<n> Experimental results demonstrate that OpenReLoc achieves superior relocalization recall and accuracy across various datasets.
  arXiv  Detail & Related papers  (2026-06-23T16:27:04Z)
• SOCO: Benchmarking Semantic Object Correspondence in Vision Foundation Models [63.28435103335999]
  We introduce SOCO, a new benchmark for Semantic Object Correspondence.<n>We show that vision foundation backbones encode strong semantic structure but transfer correspondences poorly across related categories.<n>We also show that LVLMs are stronger at text-prompted part localization than at visual-reference cross-image matching.
  arXiv  Detail & Related papers  (2026-05-29T17:58:48Z)
• From Drop-off to Recovery: A Mechanistic Analysis of Segmentation in MLLMs [50.185593677108436]
  Multimodal Large Language Models (MLLMs) are increasingly applied to pixel-level vision tasks, yet their intrinsic capacity for spatial understanding remains poorly understood.<n>We investigate segmentation capacity through a layerwise linear probing evaluation across the entire MLLM pipeline: vision encoder, adapter, and LLM.
  arXiv  Detail & Related papers  (2026-03-18T00:22:15Z)
• GameUIAgent: An LLM-Powered Framework for Automated Game UI Design with Structured Intermediate Representation [10.74393288520677]
  Game UI design requires consistent visual assets across rarity tiers yet remains a predominantly manual process.<n>We present GameUIAgent LLM-powered agentic framework that translates natural language descriptions into editable Figma designs.<n>A six-stage neuro-symbolic pipeline combines LLM generation, deterministic post-processing, and a Vision-Language Model (VLM)-guided Reflection Controller (RC) for iterative self-correction.
  arXiv  Detail & Related papers  (2026-03-16T01:57:56Z)
• Vision language models are unreliable at trivial spatial cognition [0.2902243522110345]
  Vision language models (VLMs) are designed to extract relevant visuospatial information from images.<n>We develop a benchmark dataset -- TableTest -- whose images depict 3D scenes of objects arranged on a table, and used it to evaluate state-of-the-art VLMs.<n>Results show that performance could be degraded by minor variations of prompts that use equivalent descriptions.
  arXiv  Detail & Related papers  (2025-04-22T17:38:01Z)
• Hierarchically-Structured Open-Vocabulary Indoor Scene Synthesis with Pre-trained Large Language Model [14.70850176122733]
  We propose to generate hierarchically structured scene descriptions with large language models (LLM) and then compute the scene layouts.<n>Specifically, we train a hierarchy-aware network to infer the fine-grained relative positions between objects.<n>We also present open-vocabulary scene synthesis and interactive scene design results to show the strength of our approach in the applications.
  arXiv  Detail & Related papers  (2025-02-15T05:04:14Z)
• LayoutVLM: Differentiable Optimization of 3D Layout via Vision-Language Models [57.92316645992816]
  Spatial reasoning is a fundamental aspect of human cognition, enabling intuitive understanding and manipulation of objects in three-dimensional space.<n>We introduce LayoutVLM, a framework and scene layout representation that exploits the semantic knowledge of Vision-Language Models (VLMs)<n>We demonstrate that fine-tuning VLMs with the proposed scene layout representation extracted from existing scene datasets can improve their reasoning performance.
  arXiv  Detail & Related papers  (2024-12-03T06:15:04Z)
• Locality Alignment Improves Vision-Language Models [55.275235524659905]
  Vision language models (VLMs) have seen growing adoption in recent years, but many still struggle with basic spatial reasoning errors.<n>Our goal is to resolve this with a vision backbone that effectively captures both local and global image semantics.<n>We propose a new efficient post-training stage for ViTs called locality alignment and a novel fine-tuning procedure called MaskEmbed.
  arXiv  Detail & Related papers  (2024-10-14T21:01:01Z)
• Mixed Diffusion for 3D Indoor Scene Synthesis [55.94569112629208]
  We present MiDiffusion, a novel mixed discrete-continuous diffusion model designed to synthesize plausible 3D indoor scenes.<n>We show it outperforms state-of-the-art autoregressive and diffusion models in floor-conditioned 3D scene synthesis.
  arXiv  Detail & Related papers  (2024-05-31T17:54:52Z)
• Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Models [73.40350756742231]
  Visually-conditioned language models (VLMs) have seen growing adoption in applications such as visual dialogue, scene understanding, and robotic task planning. Despite the volume of new releases, key design decisions around image preprocessing, architecture, and optimization are under-explored.
  arXiv  Detail & Related papers  (2024-02-12T18:21:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.