AI Paper Atlas

Research map / 3D vision / Gaussian splatting

MVSGaussian: Fast Generalizable Gaussian Splatting Reconstruction from Multi-View Stereo

Tianqi Liu, Guangcong Wang, Shoukang Hu, Liao Shen, Xinyi Ye, Yuhang Zang, Zhiguo Cao, Wei Li, Ziwei Liu

ECCV 2024 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting · 109 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 cascade structure (ablation)MVSGaussian · DTUPSNR: 26.71 -> 27.48Improved
add depth-aware volume rendering to splatting (hybrid rendering) (ablation)MVSGaussian · DTUPSNR: 27.48 -> 28.21Improved
add depth-aware volume rendering to splatting (hybrid rendering) (ablation)MVSGaussian · Tanks and TemplesPSNR: 21.70 -> 23.28Improved
decode RGB directly instead of SH coefficients (ablation)MVSGaussian · NeRF SyntheticPSNR: 24.27 -> 26.46Improved
decode RGB directly instead of SH coefficients (ablation)MVSGaussian · DTUPSNR: 28.19 -> 28.21No clear effect
use multi-view consistency check instead of direct concatenation for initialization (ablation)per-scene optimization · Real Forward-facingPSNR: 26.18 -> 26.98Improved
use consistency check instead of voxel downsampling (ablation)per-scene optimization · Real Forward-facingPSNR: 26.72 -> 26.98Improved
add density prediction for volume rendering (ablation)MVSGaussian · NeRF SyntheticPSNR: 26.22 -> 26.46Improved
increase input views from 2 to 4 (ablation)MVSGaussian · DTUPSNR: 25.78 -> 28.43Improved
sample 2 Gaussians per pixel instead of 1 (ablation)MVSGaussian · DTUPSNR: 28.21 -> 28.26 (memory 0.876 -> 1.508 GB)No clear effect
initialize 3D-GS from generalizable model instead of COLMAP3D-GS · Real Forward-facingPSNR / fine-tuning time: 23.92 at 10min -> 26.98 at 45sImproved
use MVS-derived pixel-aligned Gaussians instead of epipolar Transformer (PixelSplat)generalizable Gaussian Splatting · NeRF Synthetic 2-viewPSNR: 15.77 -> 25.34Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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