Semi-supervised Feature Learning For Improving Writer Identification

TL;DR

This paper introduces a semi-supervised feature learning pipeline with a novel weighted label smoothing regularization method that leverages unlabeled data to enhance offline writer identification performance.

Contribution

It proposes a new semi-supervised learning approach with WLSR for data augmentation, improving discriminative feature learning in writer identification tasks.

Findings

Significant performance improvements on ICDAR2013 and CVL datasets.

WLSR regularization enhances CNN discriminative features.

Competitive results compared to existing methods.

Abstract

Data augmentation is usually used by supervised learning approaches for offline writer identification, but such approaches require extra training data and potentially lead to overfitting errors. In this study, a semi-supervised feature learning pipeline was proposed to improve the performance of writer identification by training with extra unlabeled data and the original labeled data simultaneously. Specifically, we proposed a weighted label smoothing regularization (WLSR) method for data augmentation, which assigned the weighted uniform label distribution to the extra unlabeled data. The WLSR method could regularize the convolutional neural network (CNN) baseline to allow more discriminative features to be learned to represent the properties of different writing styles. The experimental results on well-known benchmark datasets (ICDAR2013 and CVL) showed that our proposed semi-supervised feature learning approach could significantly improve the baseline measurement and perform competitively with existing writer identification approaches. Our findings provide new insights into offline write identification.