AI Paper Atlas

Research map / 3D vision / 3D generation

Baking Gaussian Splatting into Diffusion Denoiser for Fast and Scalable Single-stage Image-to-3D Generation and Reconstruction

Yuanhao Cai, He Zhang, Kai Zhang, Yixun Liang, Mengwei Ren, Fujun Luan, Qing Liu, Soo Ye Kim, Jianming Zhang, Zhifei Zhang, Yuqian Zhou, Yulun Zhang and 3 more

ICCV 2025 · 3D vision › 3D generation · approach: Score distillation for text-to-3D · 24 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 the 3D Gaussian diffusion framework to the baseline denoiser (ablation)DiffusionGS baseline · GSOPSNR / FID: 17.63 -> 20.57 PSNR; 118.31 -> 47.86 FIDImproved
add point distribution loss Lpd for warm-up (ablation)DiffusionGS · GSOPSNR / FID: 20.57 -> 20.94 PSNR; 47.86 -> 28.41 FIDImproved
add scene-object mixed training (without RPPC) (ablation)DiffusionGS · GSOPSNR / FID: 20.94 -> 21.73 PSNR; 28.41 -> 17.79 FIDImproved
add Reference-Point Plucker Coordinate (RPPC) camera conditioning (ablation)DiffusionGS · GSOPSNR / FID: 21.73 -> 22.07 PSNR; 17.79 -> 11.52 FIDImproved
add scene-object mixed training for scene reconstruction (ablation)DiffusionGS · RealEstate10KPSNR / FID: 0.61 dB in PSNR and 10.53 in FIDImproved
add RPPC camera conditioning for scene reconstruction (ablation)DiffusionGS · RealEstate10KPSNR / FID: 0.28 dB in PSNR and 7.09 in FIDImproved
replace 2D multi-view diffusion plus post-hoc Gaussian fitting (PhotoNVS) with single-stage DiffusionGSsingle-view scene reconstruction · RealEstate10KPSNR / FID: 15.31 -> 21.63 PSNR; 28.30 -> 15.87 FIDImproved
use DiffusionGS instead of DMV3D for object generationsingle-view object generation · ABO / GSOPSNR / FID: 2.2/1.25 dB in PSNR and 23.25/21.96 in FIDImproved
use DiffusionGS instead of Flash3D (which uses a depth estimator)single-view scene reconstruction · RealEstate10KPSNR / FID: 20.29 -> 21.63 PSNR; 35.03 -> 15.87 FIDImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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