AI Paper Atlas

Research map / 3D vision / Gaussian splatting

VIRGi: View-dependent Instant Recoloring of 3D Gaussians Splats

Alessio Mazzucchelli, Ivan Ojeda-Martin, Fernando Rivas-Manzaneque, Elena Garcés, Adrián Peñate-Sánchez, Francesc Moreno-Noguer

TPAMI 2026 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS · 1 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
add multi-view rasterization training (ablation)VIRGi without decoupling · Mip-NeRF 360 recoloringPSNR: 29.19 -> 29.45Improved
add multi-view rasterization training (ablation)VIRGi without decoupling · LLFF recoloringPSNR: 28.01 -> 28.54Improved
add decoupled diffuse/specular output without multi-view (ablation)VIRGi · Mip-NeRF 360 recoloringPSNR: 29.19 -> 28.58Hurt
add decoupled output on top of multi-view (ablation)VIRGi · Mip-NeRF 360 recoloringPSNR: 29.45 -> 29.61Improved
add decoupled output on top of multi-view (ablation)VIRGi · LLFF recoloringPSNR: 28.54 -> 29.15Improved
add decoupled output on top of multi-view (ablation)VIRGi · NeRF Synthetic recoloringPSNR: 31.90 -> 31.50Hurt
feed diffuse activations into segmentation network (DS) (ablation)VIRGi · Mip-NeRF 360 recoloringPSNR: 29.61 -> 29.74Improved
feed diffuse activations into segmentation network (DS) (ablation)VIRGi · LLFF recoloringPSNR: 29.15 -> 29.10Hurt
feed diffuse activations into segmentation network (DS) (ablation)VIRGi · NeRF Synthetic recoloringPSNR: 31.50 -> 31.65Improved
train C3DGS with multi-view batches (ablation)C3DGS · Mip-NeRF 360 reconstructionPSNR: 28.96 -> 29.47Improved
train C3DGS with multi-view batches (ablation)C3DGS · NeRF Synthetic reconstructionPSNR: 35.60 -> 35.84Improved
train VIRGi with multi-view batches (ablation)VIRGi · NeRF Synthetic reconstructionPSNR: 33.87 -> 35.76Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 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