SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects
The paper introduces SceneCode, a framework designed for generating editable indoor scenes with articulated objects. It addresses limitations in existing scene synthesis methods by enabling programmatic world generation from natural language prompts. The results demonstrate improved scene generation fidelity and asset quality, making it a significant advancement in the field of embodied AI.
- ▪SceneCode formulates indoor scene synthesis as programmatic world generation.
- ▪The framework compiles natural language prompts into executable code-driven indoor worlds.
- ▪Results indicate that SceneCode enhances prompt-faithful scene generation and produces higher quality assets.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.19587 (cs) [Submitted on 19 May 2026] Title:SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects Authors:Puyi Wang, Yuhao Wang, Linjie Li, Zhengyuan Yang, Kevin Qinghong Lin, Yangguang Li, Yu Cheng View a PDF of the paper titled SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects, by Puyi Wang and 6 other authors View PDF HTML (experimental) Abstract:Indoor scene synthesis underpins embodied AI, robotic manipulation, and simulation-based policy evaluation, where a useful scene must specify not only what the environment looks like, but also how its objects are structured.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
