Research map / Label-efficient and robust learning / Recognition and long-tail
Offline Handwritten Signature Modeling and Verification Based on Archetypal Analysis
Elias N. Zois, Ilias Theodorakopoulos, George Economou
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 |
|---|---|---|---|
| use small patches (Nsp=5) instead of large patches (Nsp=19), p=30 (ablation) | archetypal analysis features · CEDAR | EER (%): 6.14 -> 2.07 | Improved |
| use small patches (Nsp=5) instead of large patches (Nsp=19), p=30 (ablation) | archetypal analysis features · MCYT75 | EER (%): 4.99 -> 3.97 | Improved |
| raise archetypes from 10 to 30 at Nsp=5 (ablation) | archetypal analysis features · CEDAR | EER (%): 3.10 -> 2.07 | Improved |
| raise archetypes from 10 to 30 at Nsp=5 (ablation) | archetypal analysis features · MCYT75 | EER (%): 6.24 -> 3.97 | Improved |
| raise archetypes from 30 to 50 at Nsp=5 (ablation) | archetypal analysis features · CEDAR | EER (%): 2.07 -> 2.26 | No clear effect |
| raise archetypes from 30 to 50 at Nsp=5 (ablation) | archetypal analysis features · MCYT75 | EER (%): 3.97 -> 4.31 | No clear effect |
| raise archetypes from 10 to 50 at Nsp=19 (ablation) | archetypal analysis features · CEDAR | EER (%): 5.36 -> 6.12 | Hurt |
| raise archetypes from 10 to 50 at Nsp=19 (ablation) | archetypal analysis features · MCYT75 | EER (%): 7.57 -> 5.46 | Improved |
| use archetypal analysis with convexity constraints instead of K-SVD sparse coding (ablation) | signature feature coding · CEDAR | FAR(S) (%): 6.83 -> 2.07 | Improved |
| use archetypal analysis instead of K-SVD sparse coding (ablation) | signature feature coding · CEDAR | FRR (%): 7.32 -> 2.07 | Improved |
| use archetypal analysis instead of K-SVD sparse coding (ablation) | signature feature coding · CEDAR | FAR(R) (%): 0.34 -> 0.17 | Improved |
| archetypes versus bag of visual words with KAZE (trained with forgeries) | signature verification · CEDAR | FRR (%): 1.60 -> 2.07 | Hurt |
Closest papers by meaning
- Static-Dynamic Interaction Networks for Offline Signature Verification (AAAI 2021 · similarity 0.84)
- Sequential Motif Profiles and Topological Plots for Offline Signature Verification (CVPR 2020 · similarity 0.84)
- SynSig2Vec: Forgery-Free Learning of Dynamic Signature Representations by Sigma Lognormal-Based Synthesis and 1D CNN (TPAMI 2021 · similarity 0.82)
- Privacy-Preserving Biometric Verification With Handwritten Random Digit String (TPAMI 2025 · similarity 0.81)
- ManiFPT: Defining and Analyzing Fingerprints of Generative Models (CVPR 2024 · similarity 0.80)
- On the Generalization of Handwritten Text Recognition Models (CVPR 2025 · similarity 0.79)
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.
