AI Paper Atlas

Research map / 3D vision / Gaussian splatting

LUDVIG: Learning-Free Uplifting of 2D Visual Features to Gaussian Splatting Scenes

Juliette Marrie, Romain Menegaux, Michael Arbel, Diane Larlus, Julien Mairal

ICCV 2025 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting · 19 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
uplift features into 3D and render instead of single-view 2D prediction (ablation)DINOv2/SAM2 segmentation · SPIn-NeRF multi-view segmentationIoU: 88.5 -> 91.6 (DINOv2); 88.6 -> 93.8 (SAM2)Improved
add 3D graph diffusion to uplifted DINOv2 features (ablation)DINOv2 segmentation · SPIn-NeRF multi-view segmentationIoU: 91.6 -> 93.8Improved
add SAM segmentation to CLIP relevancy with graph diffusion (ablation)LUDVIG CLIP segmentation · LERF object segmentationIoU overall: 50.4 -> 64.3Improved
add graph diffusion to raw CLIP relevancy (ablation)LUDVIG CLIP segmentation · LERF object segmentationIoU overall: 33.6 -> 50.4Improved
add graph diffusion on top of SAM-based segmentation (ablation)LUDVIG CLIP segmentation · LERF object segmentationIoU overall: 57.3 -> 64.3Improved
run more iterations of reconstruction-loss optimization instead of single aggregation step (ablation)Feature uplifting · feature qualityfeature qualityNo clear effect
filter out half of the Gaussians by importance (ablation)LUDVIG · all tasksqualityNo clear effect
use learning-free uplifting instead of LangSplat feature learningOpen-vocabulary segmentation · LERF 2D object segmentationruntime (min): 105 -> 10Improved
replace OpenGaussian feature training with direct upliftingOpenGaussian pipeline · ScanNet semantic segmentation (19 classes)mIoU: 24.7 -> 33.9Improved
use LUDVIG with DINOv2 and SAM masks instead of SAGA/OmniSeg3DMulti-view segmentation · SPIn-NeRF multi-view segmentationIoU: 94.3 -> 93.8 (vs OmniSeg3D)No clear effect
use purely geometric mask reprojection on 360-degree scenes (ablation)Multi-view segmentation baseline · SPIn-NeRF multi-view segmentationIoU: 93.8 -> 73.1Hurt

Later papers that compared with this method

3 later accepted papers report LUDVIG as a baseline in their results tables; 3 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 2 of 4 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