ArtVIP: Articulated Digital Assets of Visual Realism, Modular Interaction, and Physical Fidelity for Robot Learning

🎉🎉🎉 ArtVIP is accepted by ICLR 2026

---

🚀 Note
For the full dataset description, usage instructions, and downloads, please visit our 👉 Hugging Face repository.

Version History

🔮 Next Steps

• ⚡ Adapt dynamics to Isaac Sim 5.0
• 🛠️ Ensure repository maintenance for at least 2 years

Key Features

206 high-quality digital-twin articulated objects

Digital Twin Scenes

Reusable Modular Interaction

Physics Fidelity

Pixel-level Affordance Annotations

Dataset Structure

1. Articulated Objects

Click here for the introduction to Articulated objects. The specific modular-interaction types:

Damping_Effect_cabinet

Magnetic_Effect_dishwasher

Trigger_Interactions_table

Damping_Effect_cabint

Magnetic_Effect_refrigerator

Trigger_Interactions_trash_can

2. Scenes

• Digital Twin Scenes
  📁 ArtVIP/Scenes contains 6 user-defined digital twin scenes.

• Interactive Scenes
  📁 ArtVIP/Interactive_scenes contains 6 pre-configured interactive scenes.

🍳 Kitchen Scene Interaction

📺 Small Living Room Interaction

🛋️ Parlor Scene Interaction

🛏️ Bedroom Scene Interaction

---

Support

💬 Join the Community

If you're interested in ArtVIP, welcome to join our Discord community for discussions.

🌏 For users from China

Please scan the WeChat QR code below:

---

BibTeX

@inproceedings{jin2026artvip,
  title={ArtVIP: Articulated Digital Assets of Visual Realism, Modular Interaction, and Physical Fidelity for Robot Learning},
  author={Zhao Jin and Zhengping Che and Tao Li and Zhen Zhao and Kun Wu and Yuheng Zhang and others},
  booktitle={International Conference on Learning Representations (ICLR)},
  year={2026}
}