AI Paper Atlas

Research map / 3D vision / Face and shape

What Does 2D Geometric Information Really Tell Us About 3D Face Shape?

Anil Bas, William A. P. Smith

IJCV 2019 · 3D vision › Face and shape · approach: 3D morphable face model regression · 30 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 alternating least squares with separable nonlinear least squares3DMM landmark fitting · BFM synthetic, detected landmarkssurface error d_S (mm): 4.81 -> 4.21Improved
replace alternating least squares with SNLS3DMM landmark fitting · FaceWarehouse leave-one-outsurface error: about 30% lower errorImproved
fit at 120cm instead of true 30cm distance (ablation)perspective 3DMM fit · BFM perspective ambiguity, true distance 30 cmlandmark error d_L (%): 0.21 -> 0.26No clear effect
fit at 120cm instead of true 30cm distance (ablation)perspective 3DMM fit · BFM perspective ambiguity, true distance 30 cmsurface error d_S (mm): 7.23 -> 13.07Hurt
fit orthographic instead of true 240cm perspective (ablation)perspective 3DMM fit · BFM perspective ambiguity, true distance 240 cmsurface error d_S (mm): 5.37 -> 5.38No clear effect
let distance be estimated as a fitting parameterperspective least squares fit · BFM distance estimationdistance error (%): 42% mean errorHurt
compare identities at same distance vs same identity at different distanceslandmark geometry · CMDP 53 subjectslandmark error (%): 3.11 vs 2.86 and 2.83Hurt
use known subject-camera distance instead of unknown (ablation)perspective fit · CMDP identity matchingROC AUC: 0.690 -> 0.892Improved
use pretrained CNN regression instead of unknown-distance geometric fit3D face regression · CMDP identity matchingROC AUC: 0.690 -> 0.866Improved
restrict flexibility modes to plausible parameter vectors (ablation)3DMM flexibility modes · BFM 70 landmarks orthographicnumber of modes: around 50 -> 7Mixed

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