Research map / 3D vision / 3D generation
A Recipe for Generating 3D Worlds from a Single Image
Katja Schwarz, Denis Rozumny, Samuel Rota Bulò, Lorenzo Porzi, Peter Kontschieder
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 anchored panorama synthesis (duplicate input image to backside) instead of ad-hoc or sequential (ablation) | panorama generation · panorama synthesis (qualitative) | panorama coherence | Improved |
| use VLM-generated directional prompts (scene, sky, ground) instead of image caption (ablation) | panorama generation · panorama synthesis (qualitative) | duplication of input content | Improved |
| use forward-backward warping for inpainting fine-tuning data instead of forward warping (ablation) | point cloud-conditioned inpainting · ScanNet++ | PSNR | Improved |
| use forward-backward warping instead of forward warping (ablation) | point cloud-conditioned inpainting · ScanNet++ | image quality | No clear effect |
| use ControlNet inpainting instead of ViewCrafter-augmented point clouds (ablation) | 3D world pipeline · generated 3DGS renderings | downstream image quality | Improved |
| add trainable grid distortion to Gaussian Splatting (ablation) | Splatfacto 3DGS · generated 3DGS renderings | robustness and detail | Improved |
| use proposed pipeline instead of WonderJourney and DimensionX video-based baselines | single-image 3D world generation · Tanks and Temples / World Labs images | CLIP-I, NIQE, BRISQUE, Q-Align | Improved |
Closest papers by meaning
- S$^{2}$VG: 3D Stereoscopic and Spatial Video Generation via Denoising Frame Matrix (TPAMI 2026 · similarity 0.88)
- Architect: Generating Vivid and Interactive 3D Scenes with Hierarchical 2D Inpainting (NeurIPS 2024 · similarity 0.88)
- WonderWorld: Interactive 3D Scene Generation from a Single Image (CVPR 2025 · similarity 0.88)
- ViewPoint: Panoramic Video Generation with Pretrained Diffusion Models (NeurIPS 2025 · similarity 0.87)
- One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (CVPR 2024 · similarity 0.87)
- CAD: Photorealistic 3D Generation via Adversarial Distillation (CVPR 2024 · similarity 0.86)
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.
