AI Paper Atlas

Research map / Video / Video generation

ReenactGAN: Learning to Reenact Faces via Boundary Transfer

Wayne Wu, Yunxuan Zhang, Cheng Li, Chen Qian, Chen Change Loy

ECCV 2018 · Video › Video generation · approach: Diffusion-based human image animation · 209 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 the boundary latent space to the encoder-decoder reconstruction (vs plain Pix2Pix) (ablation)Pix2Pix encoder-decoder · Face reconstruction (qualitative)reconstruction qualityNo clear effect
add the target-specific boundary transformer before the decoder (ablation)ReenactGAN decoder · Face reenactment (qualitative and user study)image quality (TrueSkill) and artifactsImproved
add the shape constraint loss to the GAN loss for the transformer (ablation)Boundary transformer · Transformer loss ablation (qualitative)adherence of boundary to source poseImproved
add the cycle loss on top of GAN and shape losses (ablation)Boundary transformer · Transformer loss ablation (qualitative)quality of eyes and mouth regionsImproved
use ReenactGAN boundary-space transfer instead of CycleGAN pixel-space transferCycleGAN · User study, TrueSkill qualityquality scoreImproved
apply to unseen wild input instead of seen dataReenactGAN · User study, TrueSkill qualityquality scoreHurt
use ReenactGAN instead of Face2Face for expression transferFace2Face · Facial action unit response correlation with sourceAU response correlationImproved

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