AI Paper Atlas

Research map / 3D vision / Face and shape

ImHead: A Large-scale Implicit Morphable Model for Localized Head Modeling

Rolandos Alexandros Potamias, Stathis Galanakis, Jiankang Deng, Athanasios Papaioannou, Stefanos Zafeiriou

ICCV 2025 · 3D vision › Face and shape · approach: 3D morphable face model regression · 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
replace per-region local latents (d=1248) by a single compact global identity latent with region-specific intermediate representation (ablation)imHead · MimicMe identity reconstructionChamfer distance: 0.767 -> 0.533Improved
shrink local latents to size 8 (d=312) instead of using the compact global latent (ablation)imHead · MimicMe identity reconstructionChamfer distance: 0.533 -> 0.874Hurt
add FusionNet structural blending of local features (ablation)imHead · MimicMe identity reconstructionChamfer distance: 0.674 -> 0.533Improved
add per-region local canonical space (ablation)imHead · MimicMe identity reconstructionChamfer distance: 0.884 -> 0.533Improved
add per-region local canonical space (ablation)imHead · NPHM identity reconstructionF-score @5mm: 0.723 -> 0.898Improved
use local plus global latent (d=1344) instead of compact single latent (ablation)imHead · NPHM identity reconstructionChamfer distance: 0.494 vs 0.459Mixed
train on the large curated MimicMe+NPHM data instead of NPHM data onlyimHead · MimicMe identity reconstructionChamfer distance: 0.571 -> 0.533Improved
train on curated MimicMe data instead of NPHM onlyimHead · MimicMe expression reconstructionChamfer distance: 0.623 -> 0.592Improved
train on the curated dataset instead of NPHM onlyimHead · NPHM expression reconstructionChamfer distance: 0.508 vs 0.513No clear effect
use backward warping for expression instead of forward deformation with root findingimHead · expression fitting timeseconds: 138 -> 40Improved
use single compact latent instead of NPHM local-global latentimHead · latent sizedimension: 2176 -> 256Improved
train baseline NPHM on full curated datasetNPHM · MimicMe identity reconstructionChamfer distance: 0.618 -> 0.598Improved

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