AI Paper Atlas

Research map / 3D vision / Face and shape

Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation

Mehdi Bahri, Eimear O’ Sullivan, Shunwang Gong, Feng Liu, Xiaoming Liu, Michael M. Bronstein, Stefanos P. Zafeiriou

IJCV 2021 · 3D vision › Face and shape · approach: 3D morphable face model regression · 20 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 PointNet baseline with SMF (attention, mesh decoders, no cropping) on out-of-sample scansBaseline of Chen et al. · BU-3DFE landmark errorAvg landmark error (mm): 16.45 -> 8.72Improved
replace baseline with SMFBaseline of Chen et al. · 3DMD surface error50% quantile (mm): 3.239 -> 0.447Improved
add MeIn3D and 4DFAB data (SMF+) to SMF training (ablation)SMF · 3DMD surface error50% quantile (mm): 0.447 -> 0.407Improved
add extra datasets (SMF+) (ablation)SMF · 3DMD surface error99% quantile (mm): 1.489 -> 1.326Improved
add extra datasets (SMF+) (ablation)SMF · BU-3DFE surface error (in sample)50% quantile (mm): 0.347 -> 0.333Improved
initialize NICP with SMF registration (ablation)NICP · 3DMD landmark errorAvg landmark error (mm): 10.41 -> 10.36No clear effect
initialize NICP with SMF instead of landmarksNICP · registration runtimeseconds per scan: 45-60 s -> 20 sImproved
train SMF without BU-3DFE (out of sample) (ablation)SMF · BU-3DFE landmark errorAvg landmark error (mm): 9.07 -> 8.72No clear effect
train SMF without BU-3DFE (out of sample) (ablation)SMF · BU-3DFE surface error50% quantile (mm): 0.347 -> 0.501Hurt
SMF vs NICP with landmarks on inner-face landmark errorNICP · 3DMD landmark errorAvg Face (mm): 4.43 vs 5.57Hurt
SMF vs baseline on jaw landmarksSMF · BU-3DFE landmark errorChin (mm): 11.39 -> 38.73Hurt

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