Lucida: Parse, Generate, and Place for Composable Real-to-Sim Scene Modeling
Summary
Lucida proposes a method for composable indoor scene reconstruction that distributes requirements across parsing, generation, and placement using a VLM policy to create high-fidelity editable replicas from cluttered captures.
View Cached Full Text
Cached at: 09/01/26, 11:51 AM
Paper page - Lucida: Parse, Generate, and Place for Composable Real-to-Sim Scene Modeling
Source: https://huggingface.co/papers/2608.30821
Abstract
Lucida improves composable indoor scene reconstruction by distributing pipeline requirements across parsing, asset generation, and VLM-guided placement to achieve high-fidelity editable replicas from cluttered captures.
Composable scene modeling aims to recover a real indoor scene as complete, editable object assets arranged as observed, giving robot simulation and embodied AI a simulation-ready replica of the real environment whose objects can be manipulated individually. Existing pipelines decompose the task into three steps---parse the observations into instances, generate an asset for each, and place each asset back---but every step presumes an input that a cluttered capture rarely provides: accurate instance geometry, unoccluded views, and assets that accurately match the observations. We propose Lucida, which keeps this order but redistributes the requirements, so each step consumes only what a real capture reliably provides and precision is reached at the end of the pipeline rather than demanded at its start. Lucida parses the video into ascene graphwhose nodes carry per-instancemulti-view evidence, generates a complete asset for each instance from its evidence, and places assets withGizmoAct, aVLM policythat casts placement asmulti-turn GUI interaction, manipulating the object’s gizmo in a closed loop and deciding itself when alignment is reached. Across scene-level3D object detection,object pose estimation, andscene reconstruction, Lucida improves mAP over Boxer by 69% on R2S-Scene, raisesADD-SB@0.05 from 57.8% to 83.4% on CA-1M, and increases sceneF-Scorefrom 0.794 for SAM3D to 0.924.
View arXiv pageView PDFProject pageAdd to collection
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2608.30821 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2608.30821 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2608.30821 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
SceneMosaic: Efficient and Diverse Simulation-Ready Scene Generation via Hybrid Agentic Layout Evolution
SceneMosaic combines learned image priors with vision-language agents to efficiently generate diverse, physically valid indoor scenes, achieving a 24x speedup and improved physical validity over existing methods.
Code-as-Room: Generating 3D Rooms from Top-Down View Images via Agentic Code Synthesis
A novel MLLM-based agentic framework called Code-as-Room generates 3D indoor rooms by converting top-down images into executable Blender code through a structured execution harness with cross-stage memory to maintain context.
Roomer: Reflective Object-Grounded Model Editing and Repair for 3D Indoor Layout Synthesis
Roomer is a reflective repair framework that identifies and fixes local violations in 3D indoor layouts, using a vision-language model planner and deterministic solver, with new benchmarks Roomer-CC and Roomer-Eval.
SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation
SimFoundry is a modular system that automates real-to-sim scene construction from video, generating digital twins and affordance-preserving variations for zero-shot robot policy training, achieving strong transfer to real-world tasks and high simulation-to-real performance prediction.
Repurposing 3D Generative Model for Autoregressive Layout Generation
LaviGen is a framework that repurposes 3D generative models for autoregressive 3D layout generation, using an adapted 3D diffusion model with dual-guidance self-rollout distillation to achieve 19% higher physical plausibility and 65% faster computation than state-of-the-art methods on the LayoutVLM benchmark.