Research map / Probabilistic and scientific modeling / Molecules and proteins
DeepWriteSYN: On-Line Handwriting Synthesis via Deep Short-Term Representations
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 3 DeepWriteSYN samples generated with w_KL=0 and τ=0 to one real enrolment signature | BRNN Siamese signature verification system · BiosecurID evaluation, skilled forgeries | EER: 19.79% -> 15.32% | Helped |
| Add 3 DeepWriteSYN samples generated with w_KL=0.25 and τ=0 to one real enrolment signature | BRNN Siamese signature verification system · BiosecurID evaluation, skilled forgeries | EER: 19.79% -> 16.46% | Helped |
| Increase synthetic enrolment samples to 12 using w_KL=0 and τ=0 (ablation) | BRNN Siamese signature verification system · BiosecurID evaluation, skilled forgeries | EER: 19.79% -> 9.36% | Helped |
| Increase the number of synthetic enrolment samples, up to 12 (ablation) | DeepWriteSYN-augmented signature verification · BiosecurID evaluation, skilled forgeries | EER | Helped |
| Increase w_KL from 0 toward 1 while holding τ=0 (ablation) | DeepWriteSYN signature synthesis · DeepSignDB evaluation signatures | synthetic handwriting variability | Helped |
| Increase τ while holding w_KL fixed (ablation) | DeepWriteSYN signature synthesis · DeepSignDB evaluation signatures | synthetic handwriting variability | Helped |
| Set w_KL and τ close to 1 (ablation) | DeepWriteSYN handwriting synthesis · eBioDigitDB digits and DeepSignDB signatures | input-structure fidelity | Hurt |
| Use w_KL=0 (Autoencoder behavior) rather than a positive KL weight (ablation) | DeepWriteSYN signature synthesis · DeepSignDB evaluation signatures | synthetic handwriting variability | Hurt |
| Use τ=0 rather than a higher temperature (ablation) | DeepWriteSYN handwriting synthesis · eBioDigitDB digits and DeepSignDB signatures | synthetic handwriting variability | Hurt |
Closest papers by meaning
- DiffInk: Glyph- and Style-Aware Latent Diffusion Transformer for Text to Online Handwriting Generation (ICLR 2026 · similarity 0.86)
- AI-Sketcher : A Deep Generative Model for Producing High-Quality Sketches (AAAI 2019 · similarity 0.85)
- MetaWriter: Personalized Handwritten Text Recognition Using Meta-Learned Prompt Tuning (CVPR 2025 · similarity 0.83)
- SketchINR: A First Look into Sketches as Implicit Neural Representations (CVPR 2024 · similarity 0.82)
- SpaceGTN: A Time-Agnostic Graph Transformer Network for Handwritten Diagram Recognition and Segmentation (AAAI 2024 · similarity 0.81)
- Decoupling Layout from Glyph in Online Chinese Handwriting Generation (ICLR 2025 · similarity 0.81)
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.