AI Paper Atlas

Research map / 3D vision / Face and shape

Free-HeadGAN: Neural Talking Head Synthesis With Explicit Gaze Control

Michail Christos Doukas, Evangelos Ververas, Viktoriia Sharmanska, Stefanos P. Zafeiriou

TPAMI 2023 · 3D vision › Face and shape · approach: Speech-driven 3D facial animation · 26 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 canonical key-point estimator E_can (ablation)Free-HeadGAN · VoxCeleb reenactment (15 pairs)CSIM: 0.640 -> 0.789Improved
add canonical key-point estimator E_can (ablation)Free-HeadGAN · VoxCeleb reenactment (15 pairs)FID: 66.6 -> 53.9Improved
add canonical key-point estimator E_can (ablation)Free-HeadGAN · VoxCeleb reenactment (15 pairs)AU-H: 0.304 -> 0.351Hurt
add explicit gaze conditioning via E_gaze (ablation)Free-HeadGAN · VoxCeleb reenactment (15 pairs)AGD: 17.7 -> 13.3Improved
regress dense eye meshes instead of 3D gaze vectors (ablation)E_gaze · Gaze360AGD: 12.7 -> 10.4Improved
regress dense eye meshes instead of 3D gaze vectors (ablation)E_gaze · ColumbiaAGD: 3.84 -> 1.12Improved
regress dense eye meshes instead of 3D gaze vectors (ablation)E_gaze · MPIIGazeAGD: 4.83 -> 3.35Improved
use 2 source images instead of 1 (ablation)Free-HeadGAN · VoxCeleb self-reenactment (N-shot)FID: 35.4 -> 32.7Improved
use 8 source images instead of 1 (ablation)Free-HeadGAN · VoxCeleb self-reenactment (N-shot)FID: 35.4 -> 29.9Improved
use 8 source images instead of 1 (ablation)Free-HeadGAN · VoxCeleb self-reenactment (N-shot)PSNR: 22.16 -> 24.96Improved
replace 3DMM with sparse 3D landmarks and canonical key-pointsHeadGAN · VoxCeleb self-reenactmentFID: 36.1 -> 35.4Improved
replace 3DMM with sparse 3D landmarks and canonical key-pointsHeadGAN · VoxCeleb reenactment (15 pairs)CSIM: 0.688 -> 0.789Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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