AI Paper Atlas

Research map / 3D vision / Gaussian splatting

UniGS: Modeling Unitary 3D Gaussians for Novel View Synthesis from Sparse-view Images

Jiamin Wu, Kenkun Liu, Xiaoke Jiang, Yuan Yao, Lei Zhang

ICCV 2025 · 3D vision › Gaussian splatting · approach: Sparse-view Gaussian splatting for novel views · 3 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
model unitary 3D Gaussians in world space with a DETR-like decoder instead of per-pixel local prediction and concatenation (ablation)UniGS · GSO 4 viewsPSNR: 25.6033 -> 26.5334 (Objaverse val, UNet only vs full)Improved
initialize 3D Gaussians randomly with no constraint instead of regressing from image features (ablation)UniGS · Objaverse validationPSNR: 26.5334 -> 12.1213Hurt
random initialization inside the cone of vision (ablation)UniGS · Objaverse validationPSNR: 26.5334 -> 22.6740Hurt
add cross-view attention in the feature extractor (ablation)UniGS · Objaverse validationPSNR: 25.3923 -> 26.5334Improved
add camera modulation on queries in MVDFA (ablation)UniGS · Objaverse validationPSNR: 26.1328 -> 26.5334Improved
sample per-view points instead of shared 3D sampling points (ablation)UniGS · Objaverse validationPSNR: 25.8392 -> 26.5334Improved
increase number of Gaussians beyond 19600 (ablation)UniGS · Objaverse validationPSNRNo clear effect
raise key/value downsampling rate in SESA self-attention (ablation)UniGS · Objaverse validationmemory and PSNRMixed
UniGS vs best prior feed-forward method on GSO 4 viewsUniGS · GSO 4 fixed viewsPSNR: 26.2487 -> 30.4245 (vs LGM Large); 30.0500 (GRM)Improved
train on 4 views, test on varying number of viewsUniGS · GSOPSNRImproved

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