# SynSig2Vec: Learning Representations from Synthetic Dynamic Signatures for Real-world Verification

arXiv: https://arxiv.org/abs/1911.05358 · HTML: https://mlsanity.com/paper/1911.05358/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Nov 2019.

## Built on by (most cited first, 9 of 9)

- [Synthetic Data in Human Analysis: A Survey](https://mlsanity.com/paper/2208.09191/) (2022)
- [SLOGAN: Handwriting Style Synthesis for Arbitrary-Length and Out-of-Vocabulary Text](https://mlsanity.com/paper/2202.11456/) (2022)
- [SVC-onGoing: Signature Verification Competition](https://mlsanity.com/paper/2108.06090/) (2021)
- [ICDAR 2021 Competition on On-Line Signature Verification](https://mlsanity.com/paper/2106.00739/) (2021)
- [Letter-level Online Writer Identification](https://mlsanity.com/paper/2112.02824/) (2021)
- [Attention to Warp: Deep Metric Learning for Multivariate Time Series](https://mlsanity.com/paper/2103.15074/) (2021)
- [Introduction to Presentation Attacks in Signature Biometrics and Recent Advances](https://mlsanity.com/paper/2302.08320/) (2023)
- [Training Robust Deep Physiological Measurement Models with Synthetic Video-based Data](https://mlsanity.com/paper/2311.05371/) (2023)
- [Online Signature Verification Using Augmented Path Signature and T-Mamba](https://mlsanity.com/paper/2609.08276/) (2026)

## Abstract (arXiv)

An open research problem in automatic signature verification is the skilled forgery attacks. However, the skilled forgeries are very difficult to acquire for representation learning. To tackle this issue, this paper proposes to learn dynamic signature representations through ranking synthesized signatures. First, a neuromotor inspired signature synthesis method is proposed to synthesize signatures with different distortion levels for any template signature. Then, given the templates, we construct a lightweight one-dimensional convolutional network to learn to rank the synthesized samples, and directly optimize the average precision of the ranking to exploit relative and fine-grained signature similarities. Finally, after training, fixed-length representations can be extracted from dynamic signatures of variable lengths for verification. One highlight of our method is that it requires neither skilled nor random forgeries for training, yet it surpasses the state-of-the-art by a large margin on two public benchmarks.