AI Paper Atlas

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

Zero-Shot Paragraph-level Handwriting Imitation with Latent Diffusion Models

Martin Mayr, Marcel Dreier, Florian Kordon, Mathias Seuret, Jochen Zöllner, Fei Wu, Andreas Maier, Vincent Christlein

IJCV 2025 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic guidance · 18 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 HTR and writer-identification losses to the VAE (ablation)latent diffusion VAE · IAM paragraph reconstructionCER_P (%): 14.50 -> 3.19Improved
add HTR and writer-identification losses to the VAE (ablation)latent diffusion VAE · IAM paragraph reconstructionCER_L (%): 9.95 -> 2.29Improved
use trained compressed VAE instead of Stable Diffusion VAE (ablation)latent diffusion · IAM paragraph reconstructionCER_P (%): 5.27 -> 3.19Improved
add ranked resampling by writer identification (rank_WI) (ablation)Ours · IAM, query synth / key genuinemAP (%): 54.73 -> 61.06Improved
add ranked resampling by HTR (rank_HTR) (ablation)Ours · IAM, query synth / key genuinemAP (%): 54.73 -> 56.57Improved
add combined ranked resampling (rank_HTR+WI) (ablation)Ours · IAM, query synth / key genuinemAP (%): 54.73 -> 59.40Improved
add combined ranked resampling (rank_HTR+WI) (ablation)Ours · IAM, query synth / key genuinetop-1 (%): 50.23 -> 54.66Improved
use cosine noise scheduler instead of linear (ablation)Ours · IAM, query synth / key genuinemAP (%): 59.40 -> 64.21; CER_L 4.95 -> 7.03Mixed
remove new-line tokens from the target text (ablation)Ours · IAM, query synth / key genuineCER_L (%): 4.95 -> 34.84Hurt
remove new-line tokens from the target text (ablation)Ours · IAM, query synth / key genuinemAP (%): 59.40 -> 66.44Mixed
our paragraph model instead of VATrparagraph imitation · IAM, query synth / key genuinemAP (%): 14.40 -> 54.73Improved
our model instead of best GAN baseline on OOD dataparagraph imitation · CVL test, query synth / key genuinemAP (%): 8.37 -> 17.11Improved

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