AI Paper Atlas

Research map / 3D vision / Face and shape

Generating 3D Faces using Convolutional Mesh Autoencoders

Anurag Ranjan, Timo Bolkart, Soubhik Sanyal, Michael J. Black

ECCV 2018 · 3D vision › Face and shape · approach: 3D morphable face model regression · 598 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 the PCA model with the convolutional mesh autoencoder (8-d latent)expression/face reconstruction · interpolation experimentmean Euclidean error (mm): 1.639 -> 0.845Improved
replace the PCA model with the convolutional mesh autoencodermodel size · interpolation experimentnumber of parameters: 120,552 -> 33,856Improved
use CoMA instead of PCA on unseen expressionsface reconstruction · extrapolation experiment (bareteeth)mean Euclidean error (mm): 1.957 -> 1.376Improved
use CoMA instead of FLAME on unseen expressionsface reconstruction · extrapolation experiment (bareteeth)mean Euclidean error (mm): 2.002 -> 1.376Improved
use CoMA instead of PCA on unseen expressionsvertices within 1 mm · extrapolation experimentvertices within 1 mm error (%): 45% -> 63.8%Improved
use CoMA instead of PCA in interpolationvertices within 1 mm · interpolation experimentvertices within 1 mm error (%): 47.3% -> 72.6%Improved
replace FLAME's PCA expression space with CoMA (DeepFLAME), 2 latent dimsFLAME · expression reconstructionmean Euclidean error (mm): 0.668 -> 0.610Improved
replace FLAME's PCA expression space with CoMA (DeepFLAME), 8 latent dimsFLAME · expression reconstructionmean Euclidean error (mm): 0.477 -> 0.432Improved
replace FLAME's PCA expression space with CoMA (DeepFLAME), 16 latent dimsFLAME · expression reconstructionmean Euclidean error (mm): 0.351 -> 0.372Hurt
replace FLAME's PCA expression space with CoMA (DeepFLAME), 16 latent dimsFLAME · expression reconstructionmedian Euclidean error (mm): 0.139 -> 0.125Improved
add the variational KL loss (w_kld=0.001) to CoMA (ablation)CoMA · latent sampling with Gaussian noisesample diversity (qualitative)Improved

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