Research map / 3D vision / Gaussian splatting
DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting
Shijie Zhou, Zhiwen Fan, Dejia Xu, Haoran Chang, Pradyumna Chari, Tejas K Bharadwaj, Suya You, Zhangyang Wang, Achuta Kadambi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add GPT-4V multi-round self-refinement of the text prompt and panorama selection (ablation) | DreamScene360 panorama generation · text-to-panorama | visual quality (qualitative) | Improved |
| add semantic (DINOv2 CLS) and geometric (depth correlation) regularization on virtual views (ablation) | panoramic 3D Gaussians · novel view rendering | artifacts (qualitative) | Improved |
| initialize Gaussians from aligned monocular depth point cloud instead of random initialization (ablation) | panoramic 3D Gaussians · novel view rendering | rendering quality (qualitative) | Improved |
| use global geometric field alignment of monocular depth instead of no geometric prior (ablation) | panoramic 3D Gaussians · shifted camera views | artifacts (qualitative) | Improved |
| use panorama-based pipeline instead of LucidDreamer progressive inpainting | text-to-3D scene generation · text-to-360 scenes | BRISQUE: 51.9764 -> 38.3911 | Improved |
| use panorama-based pipeline instead of LucidDreamer | text-to-3D scene generation · text-to-360 scenes | runtime: 6min.15sec. -> 7min.20sec. | Hurt |
Later papers that compared with this method
4 later accepted papers report DreamScene360 as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- Scene Generation at Absolute Scale: Utilizing Semantic and Geometric Guidance From Text for Accurate and Interpretable 3D Indoor Scene Generation (ECCV 2026 · better on 4 of 4 shared results)
- WorldMesh: Generating Navigable Multi-Room 3D Scenes via Mesh-Conditioned Image Diffusion (ECCV 2026 · better on 3 of 3 shared results)
- Scene4U: Hierarchical Layered 3D Scene Reconstruction from Single Panoramic Image for Your Immerse Exploration (CVPR 2025 · better on 5 of 5 shared results)
- WonderTurbo: Generating Interactive 3D World in 0.72 Seconds (ICCV 2025 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LucidDreamer — LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching (CVPR 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- BloomScene: Lightweight Structured 3D Gaussian Splatting for Crossmodal Scene Generation (AAAI 2025 · similarity 0.91)
- DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation (ICLR 2024 · similarity 0.90)
- SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis (CVPR 2025 · similarity 0.89)
- Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images (CVPR 2025 · similarity 0.89)
- Generative Sparse-View Gaussian Splatting (CVPR 2025 · similarity 0.88)
- Generative Gaussian Splatting for Unbounded 3D City Generation (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
