AI Paper Atlas

Research map / Video / Video generation

StyleTalk++: A Unified Framework for Controlling the Speaking Styles of Talking Heads

Suzhen Wang, Yifeng Ma, Yu Ding, Zhipeng Hu, Changjie Fan, Tangjie Lv, Zhidong Deng, Xin Yu

TPAMI 2024 · Video › Video generation · approach: Talking face generation · 18 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 Transformer-XL with recurrence instead of standard transformer (teacher forcing) (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.566 -> 0.786Improved
use Transformer-XL instead of standard transformer with recurrence training (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.643 -> 0.786Improved
add spatial embedding transition (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.629 -> 0.786Improved
add head style discriminator (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.736 -> 0.786Improved
add head triplet loss (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.683 -> 0.786Improved
add temporal discriminator (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.710 -> 0.786Improved
use acoustic features instead of phonemes (ablation)StyleTalk++ head pose · HeadMotionSSIM: 0.593 -> 0.786Improved
use style code as decoder query (ablation)StyleTalk++ expression · MEADM-LMD: 3.651 -> 3.249Improved
use adaptive (dynamic) feedforward layer (ablation)StyleTalk++ expression · MEADM-LMD: 4.178 -> 3.249Improved
set K=8 instead of K=4 in dynamic feedforward (ablation)StyleTalk++ expression · MEADM-LMD: 3.524 -> 3.249Improved
set K=8 instead of K=16 in dynamic feedforward (ablation)StyleTalk++ expression · MEADM-LMD: 3.396 -> 3.249Improved
add expression style discriminator (ablation)StyleTalk++ expression · MEADF-LMD: 2.483 -> 2.122Improved

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