Research map / 3D vision / NeRF and radiance fields
RefSR-NeRF: Towards High Fidelity and Super Resolution View Synthesis
Xudong Huang, Wei Li, Jie Hu, Hanting Chen, Yunhe Wang
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 |
|---|---|---|---|
| Use RefSR-NeRF instead of NeRF-HR | NeRF · NeRF-LLFF ×8 | PSNR: 24.22 -> 24.88 | Improved |
| Use RefSR-NeRF instead of NeRF-HR | NeRF · NeRF-LLFF ×8 | Speed (sec): 468.4 -> 8.9 | Improved |
| Use RefSR-NeRF instead of NeRF-HR | NeRF · NeRF-LLFF ×4 | PSNR: 25.33 -> 25.37 | No clear effect |
| Use RefSR-NeRF instead of NeRF-HR | NeRF · NeRF-LLFF ×4 | LPIPS: 0.397 -> 0.398 | Hurt |
| Add the reference-guided RefSR module (ablation) | RefSR-NeRF without reference · NeRF-LLFF ×4 | PSNR: 24.93 -> 25.37 | Improved |
| Add the reference-guided RefSR module (ablation) | RefSR-NeRF without reference · NeRF-LLFF ×4 | SSIM: 0.824 -> 0.849 | Improved |
| Use DATSR instead of the proposed RefSR model (ablation) | NeRF · NeRF-LLFF ×4 | PSNR: 25.37 -> 24.34 | Hurt |
| Use TTSR instead of the proposed RefSR model (ablation) | NeRF · NeRF-LLFF ×4 | PSNR: 25.37 -> 23.43 | Hurt |
| Fine-tune DATSR for the NeRF degradation (ablation) | NeRF+DATSR · NeRF-LLFF ×4 | PSNR: 24.34 -> 24.71 | Improved |
| Fine-tune DATSR for the NeRF degradation (ablation) | NeRF+DATSR · NeRF-LLFF ×4 | SSIM: 0.826 -> 0.834 | Improved |
Closest papers by meaning
- From NeRFLiX to NeRFLiX++: A General NeRF-Agnostic Restorer Paradigm (TPAMI 2023 · similarity 0.92)
- NeRFLix: High-Quality Neural View Synthesis by Learning a Degradation-Driven Inter-Viewpoint MiXer (CVPR 2023 · similarity 0.92)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.91)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.90)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.89)
- DiSR-NeRF: Diffusion-Guided View-Consistent Super-Resolution NeRF (CVPR 2024 · similarity 0.89)
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.
