Research map / 3D vision / 3D generation
InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images
Zhengqi Li, Qianqian Wang, Noah Snavely, Angjoo Kanazawa
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 balanced GAN sampling (ablation) | InfiniteNature-Zero · LHQ test | FID: 34.9 -> 19.4 | Improved |
| add progressive trajectory growing (ablation) | InfiniteNature-Zero · LHQ test | FID: 35.3 -> 19.4 | Improved |
| add adversarial training over long virtual trajectories (ablation) | InfiniteNature-Zero · LHQ test | FID: 61.3 -> 19.4 | Improved |
| add self-supervised view synthesis loss (ablation) | InfiniteNature-Zero · LHQ test | PSNR: 26.6 -> 28.4 | Improved |
| add self-supervised view synthesis loss (ablation) | InfiniteNature-Zero · LHQ test | FID: 23.4 -> 19.4 | Improved |
| add global sky correction (ablation) | InfiniteNature-Zero · LHQ test | FID: 24.8 -> 19.4 | Improved |
| add global sky correction (ablation) | InfiniteNature-Zero · LHQ test | PSNR: 28.3 -> 28.4 | Improved |
| full approach vs naive baseline (ablation) | InfiniteNature-Zero · LHQ test | FID: 38.1 -> 19.4 | Improved |
| train on single images only vs supervised on posed video (Infinite Nature) | InfiniteNature-Zero · ACID test | FID: 32.4 -> 19.3 | Improved |
| train on single images only vs supervised on posed video (Infinite Nature) | InfiniteNature-Zero · ACID test | PSNR (3D Photos GT): 23.0 -> 23.5 | Improved |
| train on single images only vs supervised on posed video (Infinite Nature) | InfiniteNature-Zero · ACID test | SSIM (3D Photos GT): 0.83 -> 0.81 | Hurt |
| train on single images only vs supervised on posed video (Infinite Nature) | InfiniteNature-Zero · ACID test | Style x1e5: 9.37 -> 5.63 | Improved |
Closest papers by meaning
- One View Is Enough: In-the-Wild Monocular Pretraining for Novel View Generation (NeurIPS 2026 · similarity 0.86)
- Novel View Synthesis with View-Dependent Effects from a Single Image (CVPR 2024 · similarity 0.86)
- Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis (CVPR 2025 · similarity 0.86)
- Generative Inbetweening through Frame-wise Conditions-Driven Video Generation (CVPR 2025 · similarity 0.85)
- Novel View Synthesis from A Few Glimpses via Test-Time Natural Video Completion (NeurIPS 2025 · similarity 0.85)
- ZeroNVS: Zero-Shot 360-Degree View Synthesis from a Single Image (CVPR 2024 · similarity 0.85)
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.
