AI Paper Atlas

Research map / Video / Video generation

StyleHEAT: One-Shot High-Resolution Editable Talking Face Generation via Pre-trained StyleGAN

Fei Yin, Yong Zhang, Xiaodong Cun, Mingdeng Cao, Yanbo Fan, Xuan Wang, Qingyan Bai, Baoyuan Wu, Jue Wang, Yujiu Yang

ECCV 2022 · Video › Video generation · approach: Talking face generation · 202 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
use pre-trained StyleGAN feature-space warping instead of image-space refinement (Ours vs PIRenderer)StyleHEAT · Same-identity reenactmentFID: 27.14 -> 18.02Improved
StyleHEAT vs PIRenderer on cross-identity reenactmentStyleHEAT · Cross-identity reenactmentFID: 108.56 -> 91.28Improved
StyleHEAT vs PIRenderer on cross-identity identity similarityStyleHEAT · Cross-identity reenactmentCSIM: 0.4812 -> 0.4890Improved
StyleHEAT vs FOMM on same-identity identity similarity (GAN inversion loses identity)StyleHEAT · Same-identity reenactmentCSIM: 0.8121 -> 0.7475Hurt
StyleHEAT vs PIRenderer on motion accuracy (AED/APD)StyleHEAT · Same-identity reenactmentAED: 0.1073 -> 0.1151No clear effect
add calibration network to feature warping (ablation)StyleHEAT · Qualitative ablationvisual quality (teeth, closed eyes)Improved
add domain loss on calibrated feature maps (ablation)StyleHEAT · Qualitative ablationvisual quality (detail)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