AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

ToonTalker: Cross-Domain Face Reenactment

Yuan Gong, Yong Zhang, Xiaodong Cun, Fei Yin, Yanbo Fan, Xuan Wang, Baoyuan Wu, Yujiu Yang

ICCV 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Cross-domain few-shot with transformers · 22 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 analogy constraint for unpaired cross-domain training (ablation)ToonTalker · Cartoon->Real reenactmentFID: 43.353 -> 31.852Improved
add analogy constraint (ablation)ToonTalker · Real->Cartoon reenactmentFID: 29.230 -> 28.531Improved
use query transformer instead of MLP (ablation)ToonTalker · Real->Cartoon reenactmentFID: 116.740 -> 28.531Improved
use query transformer instead of MLP (ablation)ToonTalker · Cartoon->Real reenactmentFID: 31.885 -> 31.852No clear effect
add spatial feature transformation (ablation)ToonTalker generator · Cartoon->Real reenactmentFID: 41.077 -> 31.852Improved
add backward transformer (ablation)ToonTalker · Cartoon->Real reenactmentFID: 35.265 -> 31.852Improved
increase motion-loss weight lambda_m from 0 to 100 (ablation)ToonTalker · Real->Cartoon vs Cartoon->RealFID: Real->Cartoon 26.885 -> 41.936; Cartoon->Real 33.463 -> 28.166Mixed
use ToonTalker instead of finetuned DaGAN/LIA/FOMMcross-domain reenactment · Cartoon->Real reenactmentFID: 47.222 -> 31.852Improved
use ToonTalker in same-identity real-domain reconstructionwithin-domain reenactment · VoxCeleb same-identityAKD: 1.300 -> 1.335No clear effect
use ToonTalker in cross-identity real-domain reenactmentwithin-domain reenactment · Cross-identity real domainCSIM: 0.548 -> 0.526No clear effect

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