ArtVIP: Articulated Digital Assets for Robot Learning in Isaac Sim

Provider
X-Humanoid
License
Apache-2.0 (source)
Size
5,270 files
Format
JSON, PNG (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/X-Humanoid/ArtVIP)
Access
Free and public. No account with the data provider is needed.
Category
Robotics
Source
https://arxiv.org/abs/2506.04941

About ArtVIP: Articulated Digital Assets for Robot Learning in Isaac Sim

ArtVIP (ICLR 2026) is an open set of high-fidelity articulated digital-twin assets and interactive scenes for robot learning in simulation (arXiv:2506.04941). Version 1.3 contains 476 articulated objects, 6 pre-configured interactive scenes and 6 digital-twin scenes, with physically tuned joints, modular interaction behaviours (damping, magnetic and trigger effects on cabinets, dishwashers, refrigerators, tables and trash cans) and pixel-level affordance annotations. Measured 2026-11-14, Articulated_objects/ holds nine categories: Ikea_furniture, Medical_equipment, household_items, industrial_machinery, lab_items, large_furniture, major_appliances, small_appliances and small_furniture; Scenes/ and Interactive_scene/ hold the scene files. Assets are USD files intended for NVIDIA Isaac Sim 5.0 and can be inspected with the pxr USD Python bindings. The repo totals about 16 GB. Useful for sim-to-real manipulation of doors, drawers and appliances, synthetic data generation and benchmarking articulated-object policies. The card carries an Apache-2.0 licence badge; cite the ArtVIP paper.

How to load ArtVIP: Articulated Digital Assets for Robot Learning in Isaac Sim in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="X-Humanoid/ArtVIP", filename="Articulated_objects/Ikea_furniture/BRIMNES_Cabinet_with_doors_glass_black_78x95cm/NVIDIA/Materials/2023_1/Base/Glass/Clear_Glass.mdl", repo_type="dataset")
print(path)

Data structure

File types in sampled files: JSON, PNG

Files

Example file: Articulated_objects/Ikea_furniture/BRIMNES_Cabinet_with_doors_glass_black_78x95cm/NVIDIA/Materials/2023_1/Base/Glass/Clear_Glass.mdl

Analyze this dataset without downloading it

On Scigantic, ArtVIP: Articulated Digital Assets for Robot Learning in Isaac Sim opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Robotics datasets