Research map / 3D vision / NeRF and radiance fields
Neural 3D Video Synthesis From Multi-View Video
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 DyNeRF instead of NeRF-T | 10-second sequence rendering · 10-second sequence | PSNR: 28.4487 -> 29.5808 | Helped |
| add hierarchical training and importance sampling (ablation) | DyNeRF · 10-second sequence | PSNR: 28.4994 -> 29.5808 | Helped |
| add hierarchical training and importance sampling (ablation) | DyNeRF · 10-second sequence | LPIPS: 0.0985 -> 0.0832 | Helped |
| use global-median importance sampling (ablation) | DyNeRF · 10-second sequence | PSNR: 28.4994 -> 29.4623 | Helped |
| use temporal-difference importance sampling (ablation) | DyNeRF · 10-second sequence | PSNR: 28.4994 -> 29.7161 | Helped |
| combine global-median and temporal-difference sampling (ablation) | DyNeRF · 10-second sequence | LPIPS: 0.0985 -> 0.0832 | Helped |
| increase latent code dimension from 8 to 1,024 (ablation) | DyNeRF · 60 consecutive frames | PSNR: 26.4349 -> 27.6286 | Helped |
| increase latent code dimension from 1,024 to 8,192 (ablation) | DyNeRF · 60 consecutive frames | PSNR: 27.6286 -> 27.4100 | Hurt |
| use DyNeRF instead of NeuralVolumes | novel-view rendering · 10-second sequence | PSNR: 22.7975 -> 29.5808 | Helped |
| use DyNeRF instead of LLFF | novel-view rendering · 10-second sequence | PSNR: 23.2388 -> 29.5808 | Helped |
| use DyNeRF instead of NeRF-T | video perceptual quality · 10-second sequence | JOD: 7.73 -> 8.07 | Helped |
| freeze keyframe latent codes during training (ablation) | DyNeRF latent-code optimization · rendering quality | sharpness | Hurt |
Closest papers by meaning
- Streaming Radiance Fields for 3D Video Synthesis (NeurIPS 2022 · similarity 0.91)
- TeTriRF: Temporal Tri-Plane Radiance Fields for Efficient Free-Viewpoint Video (CVPR 2024 · similarity 0.89)
- Efficient-3Dim: Learning a Generalizable Single-image Novel-view Synthesizer in One Day (ICLR 2024 · similarity 0.89)
- Joint Optimization of Neural Radiance Fields and Continuous Camera Motion from a Monocular Video (CVPR 2025 · similarity 0.88)
- GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis (CVPR 2025 · similarity 0.88)
- Dynamic View Synthesis as an Inverse Problem (NeurIPS 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.