AI Paper Atlas

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

Parametric Shadow Control for Portrait Generation in Text-to-Image Diffusion Models

Haoming Cai, Tsung-Wei Huang, Shiv Gehlot, Brandon Y. Feng, Sachin Shah, Guan-Ming Su, Christopher Metzler

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic guidance · 4 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
apply shadow optimization at early/middle denoising steps instead of later steps (ablation)Shadow Director · 800 generated portraitsCVS (identity): 0.9325 -> 0.9460Improved
apply shadow optimization at early/middle denoising steps instead of last step only (ablation)Shadow Director · 800 generated portraitsCVTA: 0.2356 -> 0.2692Improved
use UNet internal features instead of UNet output as estimator input (ablation)Shadow Director · 800 generated portraitsCVS (identity): 0.8945 -> 0.9460Improved
estimate shadow in latent space instead of noisy RGB space (ablation)Shadow Director · 800 generated portraitsCLIPIQA+: 0.3220 -> 0.4484Improved
use dedicated ID-Estimator embedding instead of L1 loss on latent features (ablation)Shadow Director · 800 generated portraitsCVS (identity): 0.7685 -> 0.9460Improved
add ID-Estimator constraint to latent optimization (ablation)Shadow Director · generated portraits (qualitative)visual artifactsImproved
use test-time latent optimization during generation instead of post-hoc editing (vs IC-Light)portrait relighting · 800 generated portraitsCVS (identity): 0.9140 -> 0.9460Improved
choose Shadow Director over other relighting methods in user studyportrait relighting · User studypreference rate: 27.44% -> 45.69% (IC-Light vs ours)Improved
increase optimization iterations (shadow strength) (ablation)Shadow Director · generated portraits (qualitative)shadow intensityImproved

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