We present SSL-HV: Self-Supervised Learning approaches applied to the task of Handwriting Verification. This task involves determining whether a given pair of handwritten images originate from the same or different writer distribution. We have compared the performance of multiple generative, contrastive SSL approaches against handcrafted feature extractors and supervised learning on CEDAR AND dataset. We show that ResNet based Variational Auto-Encoder (VAE) outperforms other generative approaches achieving 76.3% accuracy, while ResNet-18 fine-tuned using Variance-Invariance-Covariance Regularization (VICReg) outperforms other contrastive approaches achieving 78% accuracy. Using a pre-trained VAE and VICReg for the downstream task of writer verification we observed a relative improvement in accuracy of 6.7% and 9% over ResNet-18 supervised baseline with 10% writer labels.
@article{arxiv.2405.18320,
title = {Self-Supervised Learning Based Handwriting Verification},
author = {Mihir Chauhan and Mohammad Abuzar Hashemi and Abhishek Satbhai and Mir Basheer Ali and Bina Ramamurthy and Mingchen Gao and Siwei Lyu and Sargur Srihari},
journal= {arXiv preprint arXiv:2405.18320},
year = {2024}
}
Comments
8 pages, 2 figures, 2 tables, Accepted at Irish Machine Vision and Image Processing Conference 2024