BoooomNing/nova-scenesmith

Text-to-Scene Environments for Embodied AI

Nova3D Website · Open Nova Studio

20 indoor simulation environments generated with Nova3D from descriptions released with SceneSmith, with editable scene programs, object and part annotations, articulated mechanisms, and native simulator exports.

SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes (ICML 2026), by Nicholas Pfaff, Thomas Cohn, Sergey Zakharov, Rick Cory, and Russ Tedrake at MIT and Toyota Research Institute, investigates natural-language generation of environments for robot simulation. This collection contributes Nova3D outputs for 20 of its released descriptions, retaining the original prompt text and scene IDs.

Nova3D generates and executes procedural code to construct the rooms, furnishings, materials, and mechanisms. Each release connects the description to its generated program, exported geometry, annotations, and simulation files, allowing researchers to inspect and modify how a scene was constructed.

The collection covers four living rooms, three dining rooms, two bedrooms, two kitchens, two recreation rooms, and seven specialized settings: an artist studio, laboratory, library, dental office, server room, pet grooming salon, and bakery.

Contributions

Contribution	Released content	Research Applications
Common-prompt scene collection	20 original descriptions with IDs linking to SceneSmith's published outputs.	Examine object coverage and spatial arrangements for the same scene request.
Procedural scene representation	Generated Python source and design specifications alongside the resulting assets.	Inspect construction logic and edit geometry, layouts, and physical parameters.
Semantic and spatial annotations	Object and part identities, labels, bounds, positions, and dimensions.	Check requested counts and relationships; measure clearance and relative placement.
Articulated and rigid-body structure	Joint pivots, axes, limits, collision geometry, and body/contact properties.	Construct opening, closing, and object-interaction experiments.
Native simulation assets	USD, MJCF, SDF, and PyBullet files with mesh and texture dependencies.	Load the same generated environment in different simulation workflows.

Inputs and correspondence

Each Nova3D generation used GPT-6 Astra and the original text description. Scene IDs retain the upstream mapping; SceneSmith geometry and images were not generation inputs. Prompt provenance and selection records the sample and source revision.

For example, scene_034 requests a dining room with a central bar table, a wall shelf holding a plant and jug, a fridge, and a wine cabinet. Its scene source and annotations let researchers inspect the resulting objects, arrangement, and mechanisms.

The common descriptions support paired inspection of Nova3D and SceneSmith outputs. A controlled method comparison additionally requires matched generation budgets and a shared evaluation protocol.

Dataset Schema

Paths are relative to this repository. Export columns identify included native files.

Column	Meaning
preview_image	Rendered scene overview.
environment	Readable description of the setting.
scene_id	Original SceneSmith scene and prompt identifier.
base_glb	Base scene geometry and appearance in GLB.
source_code	Full generated Python scene source.
isaac_sim_export	Native USD files are included.
mujoco_export	Native MuJoCo MJCF files are included.
gazebo_export	Native Gazebo SDF files are included.
inputs	Input description and upstream provenance record.
generation_model	GPT-6 Astra.
part_count	Number of pieces in the exported scene specification.
joint_count	Number of parts declaring hinge or sliding motion.
parts	Part identifiers, labels, bounds, and physical declarations.
articulation	Joint-bearing parts with pivot, axis, limits, and body properties.
measurements	Generated geometry bounds and dimensions in metres.
physics	World and body physical parameters.
materials	Texture paths and native material-binding locations.
isaac_usd	USD scene entrypoint.
mujoco_mjcf	MuJoCo XML entrypoint.
gazebo_sdf	Gazebo world entrypoint.
category	Indoor setting category.

Download

pip install huggingface_hub

Download a complete scene archive:

from huggingface_hub import hf_hub_download

archive = hf_hub_download(
    repo_id="BoooomNing/nova-scenesmith",
    repo_type="dataset",
    revision="v1.0.0",
    filename="archives/scene_034.zip",
    local_dir="nova-indoor-scenes",
)

Extract it, then open an entrypoint below. The archive contains the same scene files listed in the dataset, with all relative dependencies retained.

Tool	File inside scenes/<scene_id>/
Isaac Sim	simulation/environment.usda
MuJoCo	simulation/environment.xml
Gazebo	simulation/world.sdf
PyBullet	simulation/run_pybullet.py
GLB viewer	base/model.glb

For all 20 archives, use snapshot_download with allow_patterns=["archives/*", "LICENSE", "NOTICE", "CREDITS.md", "credits/*"]. Individual scene files are also available under scenes/.

Citation and credits

Please cite SceneSmith and this collection when using the paired descriptions and Nova3D outputs. BibTeX · Credits and licenses · Files

---

Nova3D Website · Open Nova Studio