AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

RaFE: Generative Radiance Fields Restoration

Zhongkai Wu, Ziyu Wan, Jing Zhang, Jing Liao, Dong Xu

ECCV 2024 · 3D vision › NeRF and radiance fields · approach: Dynamic and editable NeRF · 6 citations (OpenAlex)

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 changedWhereMeasuredEffect
replace per-frame NeRF fitting on restored images with a GAN-based generative two-level tri-plane NeRFRaFE · Blender 4x super-resolutionLPIPS: 0.083 -> 0.062Improved
use generative modeling instead of per-frame fittingRaFE · Blender 4x super-resolutionPSNR: 27.39 -> 24.99Hurt
use generative modeling instead of per-frame fittingRaFE · LLFF 4x super-resolutionLPIPS: 0.261 -> 0.197Improved
RaFE vs per-frame baseline on deblurringRaFE · LLFF deblurring (consistent blur)LPIPS: 0.310 -> 0.252Improved
RaFE vs per-frame baseline on denoisingRaFE · LLFF denoising (Gain 8)LPIPS: 0.281 -> 0.2561Improved
RaFE vs task-specific BAD-NeRF on pixel metric PSNRRaFE · LLFF deblurring (consistent blur)PSNR: 25.40 -> 22.93Hurt
NeRFLiX without ground-truth reference images vs RaFE on mixed degradationNeRFLiX · Blender mixed degradationLPIPS: 0.076 -> 0.107Hurt
add the residual tri-plane generator (vs directly optimizing NeRF with GAN and LPIPS loss) (ablation)RaFE · Legoimage quality metricsImproved
add the geometry (LPIPS) loss alongside GAN loss (ablation)RaFE · DrumsPSNRImproved
add GAN loss alongside perceptual loss (ablation)RaFE · Drumsperceptual metricsImproved
swap diffusion-based DiffBIR for non-diffusion SRFormer as the 2D restorer (ablation)RaFE · Blender super-resolutionperformance vs NeRF-PerframeNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy