Research map / Graph learning / Graph self-supervised learning
Sequential Motif Profiles and Topological Plots for Offline Signature Verification
Elias N. Zois, Evangelos Zervas, Dimitrios Tsourounis, George Economou
Check 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 |
|---|---|---|---|
| Represent signatures with local sequential visibility motifs and average-pool the codes | Offline signature verification · CEDAR | EER(S): 0.51 | Improved |
| Represent signatures with local sequential visibility motifs and average-pool the codes | Offline signature verification · MCYT-75 | EER(S): 1.54 | Mixed |
| Use local average-pooled motif features instead of global degree-distribution features | Visibility-graph signature verification · CEDAR | EER(S) user_thresh: 18.6 -> 0.51 | Improved |
| Use local average-pooled motif features instead of global degree-distribution features | Visibility-graph signature verification · MCYT-75 | EER(S) user_thresh: 23.3 -> 1.54 | Improved |
| Use horizontal motifs only rather than combining motif types (ablation) | Visibility motif features · CEDAR | EER(S) user_thresh: 0.51 -> 0.30 | Improved |
| Use horizontal motifs only rather than combining motif types (ablation) | Visibility motif features · MCYT-75 | EER(S) user_thresh: 1.54 -> 1.59 | No clear effect |
| Use natural motifs only rather than combining motif types (ablation) | Visibility motif features · CEDAR | EER(S) user_thresh: 0.51 -> 1.01 | Hurt |
| Use natural motifs only rather than combining motif types (ablation) | Visibility motif features · MCYT-75 | EER(S) user_thresh: 1.54 -> 2.03 | Hurt |
| Replace average pooling with max pooling (ablation) | Visibility motif features · CEDAR and MCYT-75 | EER(S) user_thresh: CEDAR: 0.51 -> 17.0; MCYT-75: 1.54 -> 12.55 | Hurt |
| Replace average pooling with standard-deviation pooling (ablation) | Visibility motif features · CEDAR and MCYT-75 | EER(S) user_thresh: CEDAR: 0.51 -> 1.99; MCYT-75: 1.54 -> 2.42 | Hurt |
| Combine average and standard-deviation pooling (ablation) | Visibility motif features · CEDAR and MCYT-75 | EER(S) user_thresh: CEDAR: 0.51 -> 0.41; MCYT-75: 1.54 -> 1.61 | Mixed |
| Increase local patch size from 5 to 7 (ablation) | Sequential motif features · CEDAR and MCYT-75 | EER(S): CEDAR: 0.04 -> 0.05; MCYT-75: 1.61 -> 1.67 | Hurt |
Closest papers by meaning
- Static-Dynamic Interaction Networks for Offline Signature Verification (AAAI 2021 · similarity 0.85)
- Offline Handwritten Signature Modeling and Verification Based on Archetypal Analysis (ICCV 2017 · similarity 0.84)
- SignGraph: A Sign Sequence is Worth Graphs of Nodes (CVPR 2024 · similarity 0.82)
- On the Power of the Weisfeiler-Leman Test for Graph Motif Parameters (ICLR 2024 · similarity 0.82)
- Linear Transformer Topological Masking with Graph Random Features (ICLR 2025 · similarity 0.81)
- Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection (NeurIPS 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.
