Research map / 3D vision / Depth and stereo
Stable View Synthesis
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use MLP Mean rather than Weighted Mean aggregation (ablation) | SVS on Tanks and Temples validation scenes · Tanks and Temples validation scenes | LPIPS%: 12.84 -> 12.51 | Helped |
| Use MLP Mean rather than MLP Max pooling (ablation) | SVS on Tanks and Temples validation scenes · Tanks and Temples validation scenes | LPIPS%: 12.65 -> 12.51 | Helped |
| Use 9 rather than 1 residual U-Net refinement stage (ablation) | SVS on Tanks and Temples validation scenes · Tanks and Temples validation scenes | LPIPS%: 12.62 -> 12.27 | Helped |
| Fine-tune network weights on the target scene (ablation) | SVS on Tanks and Temples validation scenes · Tanks and Temples validation scenes | LPIPS%: 12.19 -> 11.26 | Helped |
| Fine-tune network weights and optimize source images (ablation) | SVS on Tanks and Temples validation scenes · Tanks and Temples validation scenes | LPIPS%: 11.26 -> 9.99 | Helped |
| Apply scene-specific fine-tuning (ablation) | SVS on FVS Pirate · FVS — Pirate | LPIPS%: 29.21 -> 30.83 | Hurt |
| Apply scene-specific fine-tuning (ablation) | SVS on Tanks and Temples · Tanks and Temples — Truck, M60, Playground, Train | LPIPS%: 12.41 -> 9.34; 23.70 -> 20.45; 17.38 -> 14.22; 19.42 -> 15.73 | Helped |
| Use SVS rather than FVS | Novel view synthesis · DTU view interpolation and extrapolation | LPIPS%: 2.56/7.17 -> 1.70/5.61; 2.96/6.57 -> 1.93/3.69; 1.95/6.31 -> 1.30/4.26 | Helped |
Closest papers by meaning
- LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias (ICLR 2025 · similarity 0.89)
- MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views (NeurIPS 2024 · similarity 0.89)
- Novel View Synthesis from A Few Glimpses via Test-Time Natural Video Completion (NeurIPS 2025 · similarity 0.87)
- ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models (CVPR 2024 · similarity 0.87)
- NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis (NeurIPS 2026 · similarity 0.87)
- Video Perception Models for 3D Scene Synthesis (NeurIPS 2025 · similarity 0.87)
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.