Research map / 3D vision / 3D generation
Extreme View Synthesis
Inchang Choi, Orazio Gallo, Alejandro Troccoli, Min H. Kim, Jan Kautz
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 the refinement network after backward warping (ablation) | Extreme View Synthesis · MVS-Synth | Mean PSNR: 24.6dB -> 27.38dB | Improved |
| add the refinement network after backward warping (ablation) | Extreme View Synthesis · MVS-Synth | Mean SSIM: 0.851 -> 0.877 | Improved |
| add depth-guided input-view patches to the refinement network input (ablation) | refinement network · real scenes (qualitative) | visual quality | Improved |
| use depth probability volume pipeline instead of Stereo Magnification | Extreme View Synthesis (refined) · MVS-Synth | Mean PSNR: 25.49dB -> 27.38dB | Improved |
| use depth probability volume pipeline instead of Stereo Magnification | Extreme View Synthesis (refined) · MVS-Synth | Mean SSIM: 0.842 -> 0.877 | Improved |
| use warped (unrefined) output instead of Stereo Magnification | Extreme View Synthesis (warped) · MVS-Synth | Mean PSNR: 25.49dB -> 24.6dB | Hurt |
| blend pixels by depth probability instead of overwriting back to front (ablation) | Extreme View Synthesis · initial warping (qualitative) | refinement difficulty | Hurt |
| nearest neighbor instead of tri-linear sampling when resampling volumes (ablation) | depth probability volume resampling · resampling quality | quality / cost | No clear effect |
| increase baseline magnification from 4.5x to 30x versus Stereo Magnification | Stereo Magnification · real scenes (qualitative) | visual artifacts | Hurt |
Closest papers by meaning
- EVPGS: Enhanced View Prior Guidance for Splatting-based Extrapolated View Synthesis (CVPR 2025 · similarity 0.87)
- Novel View Synthesis from A Few Glimpses via Test-Time Natural Video Completion (NeurIPS 2025 · similarity 0.87)
- MultiDiff: Consistent Novel View Synthesis from a Single Image (CVPR 2024 · similarity 0.87)
- Novel View Synthesis with Pixel-Space Diffusion Models (CVPR 2025 · similarity 0.87)
- SimVS: Simulating World Inconsistencies for Robust View Synthesis (CVPR 2025 · similarity 0.86)
- MonoInstance: Enhancing Monocular Priors via Multi-view Instance Alignment for Neural Rendering and Reconstruction (CVPR 2025 · 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.
