AI Paper Atlas

Research map / 3D vision / Face and shape

EmoteGPT: 3D Human Facial Expression from Natural Language Descriptions

Haoran Wang, Mohit Mendiratta, Christian Theobalt, Adam Kortylewski

ECCV 2026 · 3D vision › Face and shape · approach: Speech-driven 3D facial animation

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 VQA instruction data to text-only training (ablation)EmoteGPT · Txt2Emote explicitemotion accuracy: 0.52 -> 0.56Improved
add face-image expression data on top of text+VQA (ablation)EmoteGPT · Txt2Emote explicitemotion accuracy: 0.56 -> 0.59Improved
add VQA and image data (ablation)EmoteGPT · Txt2Emote implicitarousal CCC: 0.37 -> 0.48Improved
add VQA data alone (ablation)EmoteGPT · Txt2Emote implicitvalence CCC: 0.55 -> 0.55No clear effect
use dedicated <Expr> token instead of mean pooling (ablation)EmoteGPT · Txt2Emote implicitL1 face: 1.15 -> 1.09Improved
use <Expr> token instead of <EOS> token (ablation)EmoteGPT · Txt2Emote implicitL1 face: 1.43 -> 1.09Improved
combine FLAME parameter loss with weighted mesh geometry loss (ablation)EmoteGPT · Txt2Emote implicitL1 face: 1.14 (FLAME only) / 1.13 (geo only) -> 1.09Improved
use MLLM (LLaVA-1.5) instead of CLIP regressorEmoteGPT · Txt2Emote implicitvalence CCC: 0.49 -> 0.57Improved
train without implicit descriptions, test on implicitEmoteGPT · Txt2Emote implicitvalence CCC: CLIP 0.36 vs EmoteGPT 0.49Improved
regress 3DMM parameters from text vs diffusion text-to-3D headEmoteGPT · Txt2Emote explicit (50 samples)L1 face: 1.51 (Portrait3D) -> 1.01Improved
direct text-to-3DMM vs text-to-image then fit FLAMEEmoteGPT · Txt2Emote implicitL1 face: 2.12 (Z-Image-Turbo) -> 1.09Improved
CLIP baseline vs EmoteGPT on explicit arousal CCC (explicit-only training)EmoteGPT · Txt2Emote explicitarousal CCC: CLIP 0.57 vs EmoteGPT 0.50Hurt

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