AI Paper Atlas

Research map / 3D vision / 3D generation

PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation

Cao Duy, Phong Nguyen

ECCV 2026 · 3D vision › 3D generation · approach: Score distillation for text-to-3D · 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 appearance and rotation regularization to flow matching loss (ablation)PixGS · G-Objaverse-XL ablation subsetPSNR: 16.04 -> 30.11Improved
add rotation norm regularization to appearance loss (ablation)PixGS · G-Objaverse-XL ablation subsetPSNR: NaN -> 30.11Improved
add geometry loss (ablation)PixGS · G-Objaverse-XL ablation subsetPSNR: 30.11 -> 30.24Improved
add multi-scale LoG loss (ablation)PixGS · G-Objaverse-XL ablation subsetLPIPS: 0.025 -> 0.022Improved
add ground-truth rendering supervision in phase 2 after flow matching phase 1 (ablation)PixGS · G-Objaverse-XL ablation subsetPSNR: 16.04 -> 30.36Improved
add GT-only phase 3 after phase 2 (ablation)PixGS · G-Objaverse-XL ablation subsetPSNR: 30.36 -> 31.15Improved
use View Concatenation instead of Image Prompt Adapter (ablation)PixGS · G-Objaverse-XL ablation subsetPSNR: 30.31 -> 30.36No clear effect
use View Concatenation instead of Image Prompt Adapter (ablation)PixGS · G-Objaverse-XL ablation subsetSSIM: 0.977 -> 0.969Hurt
use GSRecon pseudo-labels instead of LGM pseudo-labels (ablation)PixGS · GSOPSNR: 21.15 -> 21.21No clear effect
scale training data from 265K to 1M assetsPixGS · GSO (image-conditioned)PSNR: 20.67 -> 21.13Improved
PixGS vs DiffSplat SD-3.5 on image-conditioned generationPixGS · GSOPSNR: 19.59 -> 21.21Improved
PixGS vs best baseline on text single objectPixGS · T3Bench single objectCLIP Sim (%): 30.99 -> 31.98Improved

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