AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

FaceXFormer: A Unified Transformer for Facial Analysis

Kartik Narayan, Vibashan VS, Rama Chellappa, Vishal M. Patel

ICCV 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 58 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 MLP fusion of multi-scale features (ablation)FaceXFormer · BIWI head poseMAE: 3.70 -> 3.52Improved
add MLP fusion of multi-scale features (ablation)FaceXFormer · CelebA attributesaccuracy: 91.21 -> 91.83Improved
add MLP fusion of multi-scale features (ablation)FaceXFormer · UTKFace ageMAE: 4.21 -> 4.17Improved
add MLP fusion of multi-scale features (ablation)FaceXFormer · 300W landmarksNME: 4.70 -> 4.67Improved
add cross-attention between face and task tokens in decoder (ablation)FaceXFormer · BIWI head poseMAE: 17.16 -> 4.63Improved
add cross-attention between face and task tokens in decoder (ablation)FaceXFormer · 300W landmarksNME: 31.49 -> 5.30Improved
replace standard cross-attention with bi-directional cross-attention (ablation)FaceXFormer · BIWI head poseMAE: improvements of 1.11 MAE in head pose estimationImproved
replace standard cross-attention with bi-directional cross-attention (ablation)FaceXFormer · 300W landmarksNME: 0.63 NME in landmark detectionImproved
replace standard cross-attention with bi-directional cross-attention (ablation)FaceXFormer · CelebA attributesaccuracy: 89.98 -> 91.83Improved
replace standard cross-attention with bi-directional cross-attention (ablation)FaceXFormer · UTKFace ageMAE: 0.67 MAE in age estimationImproved
use Swin-B backbone instead of MobileNet (ablation)FaceXFormer · Backbone comparison (regression tasks)mean error: 4.64 -> 4.12Improved
use ConvNeXt-B instead of Swin-B backbone (ablation)FaceXFormer · Backbone comparison (segmentation)F1: 92.01 -> 92.08Mixed

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