AI Paper Atlas

Research map / Image generation / Editing and style transfer

SynFER: Towards Boosting Facial Expression Recognition with Synthetic Data

Xilin He, Cheng Luo, Xiaole Xian, Bing Li, Muhammad Haris Khan, Zongyuan Ge, Weicheng Xie, Siyang Song, Linlin Shen, Bernard Ghanem, Xiangyu Yue

ICCV 2025 · Image generation › Editing and style transfer · approach: Diffusion face generation and editing · 14 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 FAU control to the text-conditioned generator (ablation)SynFER generator (FEText) · Generation FER accuracyFER accuracy: 34.62 -> 48.74Improved
add FAU control to the text-conditioned generator (ablation)SynFER generator (FEText) · Generation FAU detectionFAU accuracy: 88.91 -> 92.37Improved
add semantic guidance on top of FEText+FAUs (ablation)SynFER generator · Generation FER accuracyFER accuracy: 48.74 -> 55.14Improved
add semantic guidance on top of FEText+FAUs (ablation)SynFER generator · AffectNet supervised (POSTER++)accuracy: 67.68 -> 68.13Improved
fine-tune Stable Diffusion on FEText (ablation)Stable Diffusion · AffectNet supervised (POSTER++ on synthetic data)accuracy: 65.36 -> 66.62Improved
add synthetic data to real AffectNet data for self-supervised pretrainingMoCo v3 · AffectNet linear probeaccuracy: 51.03 -> 57.84Improved
pretrain on synthetic data alone instead of real AffectNetMoCo v3 · RAF-DB linear probeaccuracy: 79.05 -> 81.17Improved
add synthetic data to real AffectNet data for pretrainingSimCLR · AffectNet linear probeaccuracy: 48.36 -> 54.37Improved
add synthetic data to real AffectNet data for pretrainingBYOL · SFEW linear probeaccuracy: 48.70 -> 51.70Improved
add synthetic data to supervised trainingPOSTER++ · AffectNetaccuracy: 67.49 -> 69.04Improved
add synthetic data to supervised trainingAPViT · RAF-DB and AffectNetaccuracy: +0.27% on RAF-DB and 0.32% on AffectNetImproved
add synthetic data to real data for few-shot trainingCDNet · RAF_C 5-way 1-shotaccuracy: 46.07 -> 46.97Improved

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