Research map / 3D vision / NeRF and radiance fields
Crowdsampling the Plenoptic Function
Zhengqi Li, Wenqi Xian, Abe Davis, Noah Snavely
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 two-phase alpha/color optimization instead of joint training from scratch (ablation) | DeepMPI full method · Trevi Fountain (Internet photo dataset) | PSNR: 21.0 -> 21.8 | Improved |
| add AdaIN conditioning instead of feeding z as input to G (ablation) | DeepMPI full method · Trevi Fountain (Internet photo dataset) | PSNR: 19.8 -> 21.8 | Improved |
| add latent features F^r to the MPI (ablation) | DeepMPI full method · Trevi Fountain (Internet photo dataset) | PSNR: 20.5 -> 21.8 | Improved |
| encode appearance with the reference-aligned deep buffer in addition to the exemplar (ablation) | DeepMPI full method · Trevi Fountain (Internet photo dataset) | PSNR: 20.9 -> 21.8 | Improved |
| add two-phase optimization (ablation) | DeepMPI full method · Sacre Coeur (Internet photo dataset) | LPIPS: 1.61 -> 1.57 | Improved |
| add AdaIN (ablation) | DeepMPI full method · Piazza Navona (Internet photo dataset) | LPIPS: 2.97 -> 2.32 | Improved |
| add latent features F^r (ablation) | DeepMPI full method · The Pantheon (Internet photo dataset) | l1 error (x10): 0.556 -> 0.495 | Improved |
| add deep buffer to encoder (ablation) | DeepMPI full method · Top of the Rock (Internet photo dataset) | LPIPS: 2.24 -> 2.48 | Hurt |
| add two-phase optimization (ablation) | DeepMPI full method · Piazza Navona (Internet photo dataset) | PSNR: 17.4 -> 17.6 | No clear effect |
| use DeepMPI full method instead of NRW baseline | NRW · Trevi Fountain (Internet photo dataset) | PSNR: 20.0 -> 21.8 | Improved |
| use DeepMPI full method instead of MUNIT baseline | MUNIT · Top of the Rock (Internet photo dataset) | LPIPS: 3.68 -> 2.48 | Improved |
Closest papers by meaning
- Neural Scene Chronology (CVPR 2023 · similarity 0.86)
- NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (ECCV 2020 · similarity 0.86)
- DecentNeRFs: Decentralized Neural Radiance Fields from Crowdsourced Images (ECCV 2024 · similarity 0.86)
- Wild-GS: Real-Time Novel View Synthesis from Unconstrained Photo Collections (NeurIPS 2024 · similarity 0.85)
- Efficient-3Dim: Learning a Generalizable Single-image Novel-view Synthesizer in One Day (ICLR 2024 · similarity 0.85)
- GenRec: Knowing Where to Reconstruct and Where to Generate (NeurIPS 2026 · 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.
