Research map / Label-efficient and robust learning / Recognition and long-tail
Anthropomorphic Features for On-Line Signatures
Moises Díaz, Miguel A. Ferrer, José J. Quintana
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 |
|---|---|---|---|
| fuse position-based and angle-based features at score level (omega=0.4) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW skilled forgeries | EER (%): 3.88 -> 3.24 | Improved |
| fuse position-based and angle-based features at score level (omega=0.4) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW random forgeries | EER (%): 0.85 -> 0.80 | No clear effect |
| use angle-based instead of position-based features | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW skilled forgeries | EER (%): 3.88 -> 3.44 | Improved |
| use real acquired pen-tip angles instead of fixed ergonomic angles (ablation) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW | EER | Hurt |
| use estimated pen-tip angles instead of real angles (ablation) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW | EER | Improved |
| fix pen-tip angles at pi/3 and 3pi/4 instead of estimated (ablation) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW | EER | Improved |
| scale trajectory ten times smaller or bigger (ablation) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW | EER | No clear effect |
| vary pen-lift handling (pz=0, pz=5mm pen-up, or wrist twist) (ablation) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW | EER | No clear effect |
| write on the ceiling instead of a horizontal table (ablation) | Virtual Skeletal Arm anthropomorphic features · MCYT-100 DTW skilled forgeries | EER | Hurt |
| use realistic per-signer bone lengths (VSAr) instead of fixed lengths | Virtual Skeletal Arm anthropomorphic features · angle features, multiple databases | EER | Improved |
| use realistic bone lengths for position features | Virtual Skeletal Arm anthropomorphic features · multiple databases | EER | No clear effect |
| use realistic bone lengths, mobile SG-NOTE Manhattan position features | Virtual Skeletal Arm anthropomorphic features · mobile SG-NOTE random forgeries | EER (%): 3.00 -> 3.67 | Hurt |
Closest papers by meaning
- A Behavioral Handwriting Model for Static and Dynamic Signature Synthesis (TPAMI 2016 · similarity 0.84)
- Authentic Hand Avatar from a Phone Scan via Universal Hand Model (CVPR 2024 · similarity 0.79)
- FSboard: Over 3 Million Characters of ASL Fingerspelling Collected via Smartphones (CVPR 2025 · similarity 0.79)
- VSNet: Focusing on the Linguistic Characteristics of Sign Language (CVPR 2025 · similarity 0.78)
- Geo-Sign: Hyperbolic Contrastive Regularisation for Geometrically Aware Sign Language Translation (NeurIPS 2025 · similarity 0.78)
- Analyzing the Synthetic-to-Real Domain Gap in 3D Hand Pose Estimation (CVPR 2025 · similarity 0.78)
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.
