AI Paper Atlas

Research map / 3D vision / Face and shape

GeoAvatar: Adaptive Geometrical Gaussian Splatting for 3D Head Avatar

SeungJun Moon, Hah Min Lew, Seungeun Lee, Ji-Su Kang, Gyeong-Moon Park

ICCV 2025 · 3D vision › Face and shape · approach: Gaussian and implicit head avatars · 6 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 Adaptive Pre-allocation Stage (rigid/flexible split) to the baseline (ablation)GaussianAvatars_0 baseline · SplattingAvatar+DynamicFace ablationLPIPS (1e-1): 0.653 -> 0.572Improved
add extended FLAME mouth structure (teeth, palate, floor) (ablation)GaussianAvatars_0 + APS · SplattingAvatar+DynamicFace ablationPSNR: 30.717 -> 32.273Improved
add part-wise mouth deformation network (ablation)GaussianAvatars_0 + APS + FLAME mouth · SplattingAvatar+DynamicFace ablationLPIPS (1e-1): 0.548 -> 0.519Improved
add L_angle regularization on the polar angle of Gaussians (ablation)GeoAvatar (without L_angle) · SplattingAvatar+DynamicFace ablationPSNR: 32.751 -> 32.697Mixed
add APS on top of FLAME mouth (versus all-flexible initialization) (ablation)GeoAvatar · SplattingAvatar+DynamicFace ablationPSNR: 31.318 -> 32.697Improved
use APS instead of treating all regions as rigid (ablation)GeoAvatar with FLAME mouth · SplattingAvatar+DynamicFace ablationLPIPS (1e-1): 0.840 -> 0.513Improved
APS or FLAME mouth alone versus together (ablation)GeoAvatar · SplattingAvatar+DynamicFace ablationPSNR: 30.356 -> 32.697Improved
train GeoAvatar on a single view versus GaussianAvatars on 16 viewsGeoAvatar · NeRSemble (12 actors)LPIPS (1e-1): 1.073 -> 0.969Improved
train GeoAvatar on single view versus GaussianAvatars on 16 views, PSNRGeoAvatar · NeRSemble (12 actors)PSNR: 27.829 -> 27.782No clear effect
use GeoAvatar instead of best baseline on DynamicFaceGeoAvatar · DynamicFacePSNR: 31.026 -> 32.760Improved
GeoAvatar costs more training time and slower inference than FlashAvatarGeoAvatar · Efficiency (RTX 3090)Inference speed (FPS): 291.20 -> 71.52Hurt

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