AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

ContraNeRF: Generalizable Neural Radiance Fields for Synthetic-to-Real Novel View Synthesis via Contrastive Learning

Hao Yang, Lanqing Hong, Aoxue Li, Tianyang Hu, Zhenguo Li, Gim Hee Lee, Liwei Wang

CVPR 2023 · 3D vision › NeRF and radiance fields · approach: Inverse rendering with neural fields · 22 citations (OpenAlex)

This paper tried a common change: Add a contrastive loss

add geometry-aware contrastive learning and cross-view attention · ScanNet (synthetic-to-real) · PSNR: 23.67 -> 24.81 Improved

What 1,317 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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
add geometry-aware contrastive learning and cross-view attentionIBRNet-style generalizable NeRF · ScanNet (synthetic-to-real)PSNR: 23.67 -> 24.81Improved
add geometry-aware contrastive learning and cross-view attentionIBRNet-style generalizable NeRF · ScanNet (synthetic-to-real)SSIM: 0.807 -> 0.831Improved
add geometry-aware contrastive learning and cross-view attentionIBRNet-style generalizable NeRF · ScanNet (synthetic-to-real)LPIPS: 0.355 -> 0.333Improved
add geometry-aware contrastive learning and cross-view attentionIBRNet-style generalizable NeRF · ScanNet (real-to-real)PSNR: 25.25 -> 25.58Improved
add geometry-aware contrastive learning and cross-view attentiongeneralizable NeRF · DTUPSNR: 26.49 -> 27.69Improved
add ContraNeRFgeneralizable NeRF · LLFFPSNR: 25.44 -> 25.44No clear effect
use ContraNeRF rather than GeoNeRFgeneralizable NeRF · NeRF SyntheticPSNR: 28.33 -> 27.92Hurt
add geometry-aware contrastive learning (ablation)BaseModel · ScanNet (synthetic-to-real)PSNR: 23.71 -> 24.53Improved
add cross-view attention (ablation)BaseModel · ScanNet (synthetic-to-real)PSNR: 23.71 -> 24.25Improved
replace geometry-aware negative sampling with random negative sampling (ablation)BaseModel with contrastive learning · ScanNet (synthetic-to-real)PSNR: 23.71 -> 23.84No clear effect
add distance-weighted contrastive loss (ablation)GeoContrast without weighting · ScanNet (synthetic-to-real)PSNR: 24.28 -> 24.53Improved
add negative pairs to GeoContrast (ablation)GeoContrast · ScanNet (synthetic-to-real)PSNR: 23.81 -> 24.81Improved

Later papers that compared with this method

1 later accepted paper reports ContraNeRF as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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