Research map / Image generation / Text-to-image diffusion
VATr++: Choose Your Words Wisely for Handwritten Text Generation
Bram Vanherle, Vittorio Pippi, Silvia Cascianelli, Nick Michiels, Frank Van Reeth, Rita Cucchiara
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add style input preparation (drop singleton punctuation style images) (ablation) | VATr++ · IAM Test, long-tail words | FID: without it +0.96 | Improved |
| add style input preparation (ablation) | VATr++ · IAM Test, all words | FID: without it -0.44 | Hurt |
| add style input preparation (ablation) | VATr++ · Balanced/IAM Test | CER: without it +0.30 | Improved |
| add text input preparation (rare-character augmentation) (ablation) | VATr++ · IAM Test | CER: without it +4.45 | Improved |
| add text input preparation (ablation) | VATr++ · IAM Test, all words | FID: without it +0.06 | No clear effect |
| add discriminator regularization (random crop) (ablation) | VATr++ · IAM Test, long-tail words | FID: without it +2.77 | Improved |
| add discriminator regularization (ablation) | VATr++ · IAM Test, all words | FID: without it +2.70 | Improved |
| add HTR model regularization (random transforms) (ablation) | VATr++ · IAM Test, long-tail words | FID: without it +2.03 | Improved |
| add HTR model regularization (ablation) | VATr++ · IAM Test | CER: without it -0.39 | Hurt |
| upgrade VATr to VATr++ | VATr++ · IAM-W16 | FID: 17.55 -> 16.29 | Improved |
| upgrade VATr to VATr++ | VATr++ · IAM-L (line-level) | FID: 22.66 -> 24.51 | Hurt |
| upgrade VATr to VATr++ | VATr++ · IAM long-tail words (TrOCR) | CER: 5.73 -> 2.34 | Improved |
Later papers that compared with this method
6 later accepted papers report VATr++ as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.
- Bridging Online and Offline Handwriting via Differentiable Physical Rendering (ECCV 2026 · better on 5 of 5 shared results)
- Learning to Generate Stylized Handwritten Text via a Unified Representation of Style, Content, and Noise (ICLR 2026 · better on 2 of 2 shared results)
- Zero-Shot Paragraph-level Handwriting Imitation with Latent Diffusion Models (IJCV 2025 · better on 6 of 6 shared results)
- Beyond Isolated Words: Diffusion Brush for Handwritten Text-Line Generation (ICCV 2025 · better on 4 of 4 shared results)
- One-Shot Diffusion Mimicker for Handwritten Text Generation (ECCV 2024 · better on 3 of 3 shared results)
- DiffusionPen: Towards Controlling the Style of Handwritten Text Generation (ECCV 2024 · better on 1 of 1 shared result)
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
- Zero-Shot Styled Text Image Generation, but Make It Autoregressive (CVPR 2025 · similarity 0.88)
- DiffusionPen: Towards Controlling the Style of Handwritten Text Generation (ECCV 2024 · similarity 0.88)
- GANwriting: Content-Conditioned Generation of Styled Handwritten Word Images (ECCV 2020 · similarity 0.88)
- On the Generalization of Handwritten Text Recognition Models (CVPR 2025 · similarity 0.85)
- Rich Human Feedback for Text-to-Image Generation (CVPR 2024 · similarity 0.84)
- Exploring the Deep Fusion of Large Language Models and Diffusion Transformers for Text-to-Image Synthesis (CVPR 2025 · similarity 0.84)
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.
