Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.16825 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912721873666048 |
|---|---|
| author | Wang, Dilin Jung, Hyunyoung Monnier, Tom Sohn, Kihyuk Zou, Chuhang Xiang, Xiaoyu Yeh, Yu-Ying Liu, Di Huang, Zixuan Nguyen-Phuoc, Thu Fan, Yuchen Oprea, Sergiu Wang, Ziyan Shapovalov, Roman Sarafianos, Nikolaos Groueix, Thibault Toisoul, Antoine Dhar, Prithviraj Chu, Xiao Chen, Minghao Park, Geon Yeong Gupta, Mahima Azziz, Yassir Ranjan, Rakesh Vedaldi, Andrea |
| author_facet | Wang, Dilin Jung, Hyunyoung Monnier, Tom Sohn, Kihyuk Zou, Chuhang Xiang, Xiaoyu Yeh, Yu-Ying Liu, Di Huang, Zixuan Nguyen-Phuoc, Thu Fan, Yuchen Oprea, Sergiu Wang, Ziyan Shapovalov, Roman Sarafianos, Nikolaos Groueix, Thibault Toisoul, Antoine Dhar, Prithviraj Chu, Xiao Chen, Minghao Park, Geon Yeong Gupta, Mahima Azziz, Yassir Ranjan, Rakesh Vedaldi, Andrea |
| contents | We introduce WorldGen, a system that enables the automatic creation of large-scale, interactive 3D worlds directly from text prompts. Our approach transforms natural language descriptions into traversable, fully textured environments that can be immediately explored or edited within standard game engines. By combining LLM-driven scene layout reasoning, procedural generation, diffusion-based 3D generation, and object-aware scene decomposition, WorldGen bridges the gap between creative intent and functional virtual spaces, allowing creators to design coherent, navigable worlds without manual modeling or specialized 3D expertise. The system is fully modular and supports fine-grained control over layout, scale, and style, producing worlds that are geometrically consistent, visually rich, and efficient to render in real time. This work represents a step towards accessible, generative world-building at scale, advancing the frontier of 3D generative AI for applications in gaming, simulation, and immersive social environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_16825 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WorldGen: From Text to Traversable and Interactive 3D Worlds Wang, Dilin Jung, Hyunyoung Monnier, Tom Sohn, Kihyuk Zou, Chuhang Xiang, Xiaoyu Yeh, Yu-Ying Liu, Di Huang, Zixuan Nguyen-Phuoc, Thu Fan, Yuchen Oprea, Sergiu Wang, Ziyan Shapovalov, Roman Sarafianos, Nikolaos Groueix, Thibault Toisoul, Antoine Dhar, Prithviraj Chu, Xiao Chen, Minghao Park, Geon Yeong Gupta, Mahima Azziz, Yassir Ranjan, Rakesh Vedaldi, Andrea Computer Vision and Pattern Recognition Artificial Intelligence We introduce WorldGen, a system that enables the automatic creation of large-scale, interactive 3D worlds directly from text prompts. Our approach transforms natural language descriptions into traversable, fully textured environments that can be immediately explored or edited within standard game engines. By combining LLM-driven scene layout reasoning, procedural generation, diffusion-based 3D generation, and object-aware scene decomposition, WorldGen bridges the gap between creative intent and functional virtual spaces, allowing creators to design coherent, navigable worlds without manual modeling or specialized 3D expertise. The system is fully modular and supports fine-grained control over layout, scale, and style, producing worlds that are geometrically consistent, visually rich, and efficient to render in real time. This work represents a step towards accessible, generative world-building at scale, advancing the frontier of 3D generative AI for applications in gaming, simulation, and immersive social environments. |
| title | WorldGen: From Text to Traversable and Interactive 3D Worlds |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2511.16825 |