Overview

3D scene generation studies how to turn prompts and visual observations into environments with coherent layouts, geometry, appearance, and sometimes motion or interaction. Methods range from procedural rules and language-model agents to diffusion, flow-matching, and autoregressive models that generate meshes, Gaussian splats, voxels, or occupancy representations. Work targets indoor design, large-scale worlds, driving scenes, and simulation for robotics, using constraints, rendering feedback, and physics checks to improve control and usability. A central challenge is balancing visual realism and scalable generation with consistent geometry, physical plausibility, editability, and reliable evaluation.

Latest papers

CardsList

200 papers12 papers in the last four weeks, down 54% on the four weeks before. 0.1% of all new papers.

  1. World-delta: Continually Integrating Observations for Incremental 3D Scene Generation

    Nov 19, 2026Haonan Wang, Hanyu Zhou, Xiaobin Hu +13D Scene Generation3D Diffusion Models

  2. GTA: Generative Tower of Agents for Scalable Holonic Scene Generation

    Nov 18, 2026Zeng Tao, Yunuo Chen, Xiang Feng +73D Scene Generation4D Scene Generation

  3. SceneAgent: A Multi-Agent Framework for Portable Game Scene Generation

    Nov 13, 2026Xingyi Li, Jiahao Yang, Yiqin Zhu +3Game Generation3D Scene Generation

  4. ReconDreamer#: Self-Evolving Closed-Loop Synergy of Reconstruction and Generation for Dynamic Driving Scenes

    Nov 3, 2026Songling Liu, Guosheng Zhao, Xiaofeng Wang +5Dynamic Scene Reconstruction3D Scene Generation

  5. TRAILS: Trajectory Generation for Automated Indoor Locomotion Studies

    Nov 2, 2026Juho Kalliokoski, Evan G. Center, Steven M. LaValle +2Procedural Material GenerationTrajectory Generation

  6. VoxLution: Evolving Voxels for High-Fidelity 3D Scene Generation

    Oct 30, 2026Daoyi Gao, Xinyi Zhang, Alexey Bokhovkin +23D Scene GenerationIndoor Scene Generation

  7. LucidLyra: Feed-Forward Generative 3D Scene Reconstruction and Joint Understanding via Multi-Task Distillation

    Oct 29, 2026Oliver Hahn, Rodrigo Marcuzzi, Sangeetha Grama Srinivasan +73D Scene Understanding3D Scene Generation

  8. DyCraft: An Agentic System for Long-Horizon 4D Scene Generation

    Oct 29, 2026Jeff Tan, Hang Dai, Zeqian Long +54D Scene Generation3D Scene Generation

  9. ScIntBench: Physical Interaction Plausibility in 3D Scene Generation

    Oct 27, 2026Wendi Ying, Daochang Liu, Leonid Sigal +13D Scene GenerationIndoor Scene Generation

  10. CityLift: Compositional 3D Urban Generation from a Single Image

    Oct 23, 2026Haoran Gong, Jie Feng, Chenxu Wang3D Scene GenerationImage-to-3D Scene Generation

  11. WorldStream: Streaming Interactive 3D Worlds with Hierarchical Planning and Persistent State

    Oct 20, 2026Xiaoxiao Ma, Weiyang Jin, Yanwen Ma +11Interactive World Models3D Scene Generation

  12. CityPaint: A Free Lunch from 2D Diffusion for City-Scale 3D Texturing

    Oct 20, 2026Yuchen Zou, Lihuang Fang, Siyu Zhu +13D Texture Generation3D Scene Generation

  13. InfiniDream: Open-Ended Photorealistic 3D Worlds Through Procedural Generation

    Oct 19, 2026Yi-Chuan Huang, Hao-Jen Chien, Zheng-Hui Huang +3Procedural 3D Generation3D Scene Generation

  14. Scene-Designer: Learning Design Intelligence for Agentic 3D Scene Generation

    Oct 16, 2026Xuanyu Liu, Hongyan Feng, Miao Pan +53D Scene GenerationIndoor Scene Generation

  15. TriFold: Folding irregular 3D Gaussians into structured latents for 3D Scene Generation

    Oct 16, 2026Yun Wang, Chiyu Wang, Fan Jiang +53D Scene Generation3D Representation Learning

  16. PhySGA: A Physics-Space Grounded Agentic Framework for Generalizable and Interactive 3D Composition

    Oct 14, 2026Weichen Lyu, Chong Xia, Mingyu Dong +33D Scene GenerationRobotic Assembly

  17. BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References

    Oct 14, 2026Xinyu Fang, Yuchi Wang, Weihao Yu +123D Scene Generation3D Reconstruction Benchmarks

  18. SceneAE: Scene Generation and Reconstruction in Queryable 3D-Anchored Latent Space

    Oct 14, 2026Shuofeng Sun, Wenzhao Zheng, Yunpeng Zhang +33D Scene Generation3D Scene Representation

  19. CityCode: Building Agent-Operable 3D Cities from Sparse Aerial Observations

    Oct 13, 2026Jiahui Sun, Ran Bi, Yifei Liu +8Aerial 3D Reconstruction3D Scene Representation

  20. ExeWorldBench: Benchmarking Code-Based Executable 3D World Modeling

    Oct 10, 2026Xinge Peng, Zhen Fang, Ruibing Wang +9World ModelsWorld Model Evaluation

  21. Building Rome from a Single Image

    Oct 6, 2026Jiraphon Yenphraphai, Fang Li, Tianshuo Xu +53D Scene GenerationSingle-View 3D Reconstruction

  22. Controllable and Photorealistic Pedestrian Risky Motion Generation for End-to-End Driving Safety Evaluation

    Oct 5, 2026Siyuan Liu, Miao Li, Haibao Yu +4Pedestrian Behavior ModelingEnd-to-End Autonomous Driving

  23. Code2Games: Enabling Coding Agents for Gaming World Generation

    Oct 4, 2026Wei Wu, Ziyang Xu, Zeyu Zhang +2Agentic Code Generation3D Scene Generation

  24. DistScene: Object-to-Scene Distillation for 3D Scene Generation

    Oct 3, 2026Kunming Luo, Hongyu Yan, Ken Deng +63D Scene GenerationImage-to-3D Scene Generation

  25. SPHERE: Adaptive VR Indoor Scene Generation via LLM-Enhanced Spatial Preference Learning and Human-in-the-Loop RL

    Oct 1, 2026Hyeonmin Lee, Zheng Wei, Kyungmin Kwon +3Human-in-the-Loop AI3D Scene Generation