Research map / Detection and segmentation / CNN features
DetailSemNet: Elevating Signature Verification through Detail-Semantic Integration
Meng-Cheng Shih, Tsai-Ling Huang, Yu-Heng Shih, Hong-Han Shuai, Hsuan-Tung Liu, Yi-Ren Yeh, Ching-Chun Huang
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 SemanticsAttend branch (ablation) | DetailSemNet · BHSig-H | EER (%): 21.06 -> 2.07 | Improved |
| add SemanticsAttend branch (ablation) | DetailSemNet · ChiSig | EER (%): 29.01 -> 5.85 | Improved |
| increase number of reference signatures from 1 to 3 (ablation) | DetailSemNet · GPDS Synthetic | EER (%): 11.69 -> 8.06 | Improved |
| add cross-entropy classifier on siamese features (ablation) | DetailSemNet · GPDS Synthetic | EER (%): 11.69 -> 8.92 | Improved |
| set fusion weight lambda0 to 1.0 instead of 0.1, 10 or 100 (ablation) | DetailSemNet · BHSig-H | EER (%): 2.52 / 2.77 / 3.47 -> 2.07 | Improved |
| fine-tune only Conv_fuse on target dataset instead of zero-shot (ablation) | DetailSemNet · BHSig-B (trained on BHSig-H) | EER (%): 7.46 -> 4.70 | Improved |
| use ResNet50 CNN backbone instead of ViT (ablation) | DetailSemNet · ChiSig | Accuracy (%): 87.79 -> 87.28 | Hurt |
Closest papers by meaning
- SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models (CVPR 2026 · similarity 0.79)
- Static-Dynamic Interaction Networks for Offline Signature Verification (AAAI 2021 · similarity 0.77)
- Echoes within the Reasoning: Stealthy and Effective Watermarking via Chain of Thought (ICML 2026 · similarity 0.76)
- AttnDiff: Attention-based Differential Fingerprinting for Large Language Models (NeurIPS 2026 · similarity 0.76)
- Unsupervised Pretraining for Fact Verification by Language Model Distillation (ICLR 2024 · similarity 0.76)
- Beyond Text: Fine-Grained Multi-Modal Fact Verification with Hypergraph Transformers (AAAI 2025 · similarity 0.76)
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.
