AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

FairHuman: Boosting Hand and Face Quality in Human Image Generation with Minimum Potential Delay Fairness in Diffusion Models

Yuxuan Wang, Tianwei Cao, Huayu Zhang, Zhongjiang He, Kongming Liang, Zhanyu Ma

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Multi-subject personalization · 3 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 hand and face region losses to the global diffusion loss (fixed weights) (ablation)FairHuman LoRA · Human validation set (LoRA)FID: 49.48 -> 47.22Improved
add MPD fairness weighting instead of fixed-weight scalarization (ablation)FairHuman LoRA · Human validation set (LoRA)FID: 47.22 -> 46.11Improved
add regional losses to ControlNet (ablation)FairHuman ControlNet · Human validation set (ControlNet)pose mean difference: 3.45 -> 2.78Improved
add regional losses without fairness to ControlNet (ablation)FairHuman ControlNet · Human validation set (ControlNet)FID: 46.18 -> 46.79Hurt
add MPD fairness weighting to ControlNet multi-objective training (ablation)FairHuman ControlNet · Human validation set (ControlNet)pose mean difference: 2.78 -> 2.35Improved
increase LoRA rank from 64 to 256 (ablation)FairHuman LoRA · Human validation setFID: 47.87 -> 46.11Improved
use MPD fairness instead of other multi-objective strategies (DWA, UW, RLW, SI, Nash-MTL, LS) (ablation)FairHuman LoRA · Human validation setHPS (%): 32.73 vs 32.52 to 32.63Mixed
use single fairness-weighted LoRA instead of MoLE two-LoRA mixtureFairHuman LoRA · SDXL backboneextra memory usage: 2611.75MB -> 1413.12MBImproved
apply method to ControlNet instead of using ControlNet-UnionControlNet · Human validation setHPS (%): 32.80 vs 32.72No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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