AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

Charge: A Comprehensive Novel View Synthesis Benchmark and Dataset to Bind Them All

Michal Nazarczuk, Thomas Tanay, Arthur Moreau, Zhensong Zhang, Eduardo Pérez-Pellitero

CVPR 2026 · 3D vision › NeRF and radiance fields · approach: Inverse rendering with neural fields · 0 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
increase training views from 3 to 94DGS · Charge SparsePSNR: 19.71 -> 26.67Improved
increase training views from 3 to 64DGS · Charge SparsePSNR: 19.71 -> 23.93Improved
evaluate on dynamic regions instead of static regions4DGS · Charge DensePSNR: 31.85 -> 26.84Hurt
align target cameras using target views instead of Umeyama on source viewsWorldMirror · Charge static Dense NVSPSNR: 20.10 -> 24.44Improved
align target cameras using target views instead of Umeyama on source viewsWorldMirror · Charge static Sparse-3 NVSPSNR: 17.54 -> 20.94Improved
use WorldMirror instead of AnySplatWorldMirror · Charge static Sparse-3 NVSPSNR: 12.15 -> 17.54Improved
use Spline Fast trajectory instead of Random Walk Slow4DGS · Charge MonoPSNR: 23.38 -> 24.85Improved
use MoSca instead of 4DGS on Random Walk Slow trajectoryMoSca · Charge Mono (Random Walk Slow)PSNR: 23.38 -> 24.29Improved
use MoSca instead of D-3DGS on Spline Fast trajectoryMoSca · Charge Mono (Spline Fast)PSNR: 24.88 -> 23.39Hurt
use pi3 instead of VGGT for camera pose at Sparse-3pi3 · Charge static Sparse-3AUC@5: 0.3581 -> 0.1900Hurt
use pi3 instead of VGGT for depth in Densepi3 · Charge static DenseAbsRel: 0.1085 -> 0.1344Mixed
use Ex4DGS instead of 4DGS in dense setupEx4DGS · Charge DensePSNR: 28.94 -> 29.75Improved

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