SceneReVis: A Self-Reflective Vision-Grounded Framework for 3D Indoor Scene Synthesis via Multi-turn RL

• URL: http://arxiv.org/abs/2602.09432v1
• Date: Tue, 10 Feb 2026 05:55:56 GMT
• Title: SceneReVis: A Self-Reflective Vision-Grounded Framework for 3D Indoor Scene Synthesis via Multi-turn RL
• Abstract summary: Current one-pass 3D scene synthesis methods often suffer from spatial hallucinations, such as collisions, due to a lack of deliberative reasoning.<n>We introduce SceneReVis, a vision-grounded self-reflection framework that employs an iterative diagnose-and-act'' loop to explicitly intercept and resolve spatial conflicts.<n>We propose a two-stage training recipe that transitions from Supervised Fine-Tuning to Agentic Reinforcement Learning, evolving the model into an active spatial planner.
• Score: 18.967658906855466
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Current one-pass 3D scene synthesis methods often suffer from spatial hallucinations, such as collisions, due to a lack of deliberative reasoning. To bridge this gap, we introduce SceneReVis, a vision-grounded self-reflection framework that employs an iterative ``diagnose-and-act'' loop to explicitly intercept and resolve spatial conflicts using multi-modal feedback. To support this step-wise paradigm, we construct SceneChain-12k, a large-scale dataset of causal construction trajectories derived through a novel reverse engineering pipeline. We further propose a two-stage training recipe that transitions from Supervised Fine-Tuning to Agentic Reinforcement Learning, evolving the model into an active spatial planner. Extensive experiments demonstrate that SceneReVis achieves state-of-the-art performance in high-fidelity generation and goal-oriented optimization, with robust generalization to long-tail domains.

Related papers

• Dynamic-Robust Photometric-Semantic Reconstruction for Open-Vocabulary 3D Scene Understanding [63.599342750566485]
  SPAR is a novel joint semantic-geometric encoding architecture that explicitly isolates transient dynamic noise prior to latent space aggregation.<n>Our end-to-end approach achieves exceptional novel view synthesis quality, yielding a PSNR of 22.15 dB and 23.33 dB given only 3 and 4 input views respectively.
  arXiv  Detail & Related papers  (2026-08-29T09:58:24Z)
• ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoning [26.96786747869045]
  We propose ThinkBLOX, a progressive reasoning framework that iteratively designs and refines 3D scenes.<n>ThinkBLOX treats layout generation as a state-conditioned, step-by-step reasoningand-action process.<n>We show that ThinkBLOX significantly outperforms recent one-shot and iterative baselines in physical plausibility, semantic alignment, and interactive editability.
  arXiv  Detail & Related papers  (2026-07-15T07:42:47Z)
• InfiniVerse: Occupancy Guided Unbounded Scene Generation for Autonomous Driving [70.7737035270887]
  We introduce InfiniVerse, a unified pipeline for synthesis of dynamic urban scenes from a single frame.<n>InfiniVerse achieves state-of-the-art performance with a FID of 6.4 and FVD of 67.97, significantly outperforming existing benchmarks in both duration and stability.
  arXiv  Detail & Related papers  (2026-06-30T04:08:22Z)
• A Principled Approach for Creating High-fidelity Synthetic Demonstrations for Imitation Learning [1.5469452301122175]
  3D Gaussian Splatting (3DGS) has enabled visually realistic demonstration generation from a single expert trajectory and a short multi-view scan.<n>Existing 3DGS-based synthesis pipelines typically generate new motions using sampling-based planners or trajectory optimization.<n>We propose a framework that synthesizes diverse task demonstrations while explicitly preserving expert motion structure.
  arXiv  Detail & Related papers  (2026-05-02T04:09:07Z)
• Stepper: Stepwise Immersive Scene Generation with Multiview Panoramas [101.58587238272462]
  We present Stepper, a unified framework for text-driven immersive 3D scene synthesis.<n>Stepper is trained on a new large-scale, multi-view panorama dataset.<n>It achieves state-of-the-art fidelity and structural consistency, outperforming prior approaches.
  arXiv  Detail & Related papers  (2026-03-30T20:26:28Z)
• One-Shot Refiner: Boosting Feed-forward Novel View Synthesis via One-Step Diffusion [57.824020826432815]
  We present a novel framework for high-fidelity novel view synthesis (NVS) from sparse images.<n>We design a Dual-Domain Detail Perception Module, which enables handling high-resolution images without being limited by the ViT backbone.<n>We develop a feature-guided diffusion network, which can preserve high-frequency details during the restoration process.
  arXiv  Detail & Related papers  (2026-01-20T17:11:55Z)
• UniSplat: Unified Spatio-Temporal Fusion via 3D Latent Scaffolds for Dynamic Driving Scene Reconstruction [26.278318116942526]
  We present UniSplat, a feed-forward framework that learns robust dynamic scene reconstruction through unified latent-temporal fusion.<n>Experiments on real-world datasets demonstrate that UniSplat achieves state-of-the-art synthesis in novel view, while providing robust and high-quality renderings for viewpoints outside the original camera coverage.
  arXiv  Detail & Related papers  (2025-11-06T17:49:39Z)
• Follow My Hold: Hand-Object Interaction Reconstruction through Geometric Guidance [61.41904916189093]
  We propose a novel diffusion-based framework for reconstructing 3D geometry of hand-held objects from monocular RGB images.<n>We use hand-object interaction as geometric guidance to ensure plausible hand-object interactions.
  arXiv  Detail & Related papers  (2025-08-25T17:11:53Z)
• SuperGS: Consistent and Detailed 3D Super-Resolution Scene Reconstruction via Gaussian Splatting [6.309174895120047]
  3D Gaussian Splatting (3DGS) has excelled in novel view synthesis (NVS) with its real-time rendering capabilities and superior quality.<n>However, it encounters challenges for high-resolution novel view synthesis (HRNVS) due to the coarse nature of primitives derived from low-resolution input views.<n>We propose SuperGS, an expansion of Scaffold-GS designed with a two-stage coarse-to-fine training framework.
  arXiv  Detail & Related papers  (2025-05-24T11:33:57Z)
• SpatialCrafter: Unleashing the Imagination of Video Diffusion Models for Scene Reconstruction from Limited Observations [44.53106180688135]
  This work takes on the challenge of reconstructing 3D scenes from sparse or single-view inputs.<n>We introduce SpatialCrafter, a framework that leverages the rich knowledge in video diffusion models to generate plausible additional observations.<n>Through a trainable camera encoder and an epipolar attention mechanism for explicit geometric constraints, we achieve precise camera control and 3D consistency.
  arXiv  Detail & Related papers  (2025-05-17T13:05:13Z)
• Learning to Align and Refine: A Foundation-to-Diffusion Framework for Occlusion-Robust Two-Hand Reconstruction [50.952228546326516]
  Two-hand reconstruction from monocular images faces persistent challenges due to complex and dynamic hand postures.<n>Existing approaches struggle with such alignment issues, often resulting in misalignment and penetration artifacts.<n>We propose a dual-stage Foundation-to-Diffusion framework that precisely align 2D prior guidance from vision foundation models.
  arXiv  Detail & Related papers  (2025-03-22T14:42:27Z)
• OccScene: Semantic Occupancy-based Cross-task Mutual Learning for 3D Scene Generation [64.77193457268886]
  OccScene integrates fine-grained 3D perception and high-quality generation in a unified framework.<n>OccScene generates new and consistent 3D realistic scenes only depending on text prompts.<n>Experiments show that OccScene achieves realistic 3D scene generation in broad indoor and outdoor scenarios.
  arXiv  Detail & Related papers  (2024-12-15T13:26:51Z)
• Omni-Scene: Omni-Gaussian Representation for Ego-Centric Sparse-View Scene Reconstruction [9.116550622312362]
  In autonomous driving scenarios, a more practical paradigm is ego-centric reconstruction, which is characterized by minimal cross-view overlap.<n>This paper conducts an in-depth analysis of different representations, and introduces Omni-Gaussian representation with tailored network design.<n> Experiments show that our method significantly surpasses state-of-the-art methods, pixelSplat and MVSplat, in ego-centric reconstruction.
  arXiv  Detail & Related papers  (2024-12-09T07:48:15Z)
• T-3DGS: Removing Transient Objects for 3D Scene Reconstruction [83.05271859398779]
  Transient objects in video sequences can significantly degrade the quality of 3D scene reconstructions.<n>We propose T-3DGS, a novel framework that robustly filters out transient distractors during 3D reconstruction using Gaussian Splatting.
  arXiv  Detail & Related papers  (2024-11-29T07:45:24Z)
• VortSDF: 3D Modeling with Centroidal Voronoi Tesselation on Signed Distance Field [5.573454319150408]
  We introduce a volumetric optimization framework that combines explicit SDF fields with a shallow color network, in order to estimate 3D shape properties over tetrahedral grids. Experimental results with Chamfer statistics validate this approach with unprecedented reconstruction quality on various scenarios such as objects, open scenes or human.
  arXiv  Detail & Related papers  (2024-07-29T09:46:39Z)
• Unifying Correspondence, Pose and NeRF for Pose-Free Novel View Synthesis from Stereo Pairs [57.492124844326206]
  This work delves into the task of pose-free novel view synthesis from stereo pairs, a challenging and pioneering task in 3D vision. Our innovative framework, unlike any before, seamlessly integrates 2D correspondence matching, camera pose estimation, and NeRF rendering, fostering a synergistic enhancement of these tasks.
  arXiv  Detail & Related papers  (2023-12-12T13:22:44Z)
• Learning to Render Novel Views from Wide-Baseline Stereo Pairs [26.528667940013598]
  We introduce a method for novel view synthesis given only a single wide-baseline stereo image pair. Existing approaches to novel view synthesis from sparse observations fail due to recovering incorrect 3D geometry. We propose an efficient, image-space epipolar line sampling scheme to assemble image features for a target ray.
  arXiv  Detail & Related papers  (2023-04-17T17:40:52Z)

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.