Self-Training of Handwritten Word Recognition for Synthetic-to-Real Adaptation
This work addresses the challenge of limited labeled data in handwritten word recognition, offering a practical solution for applications where annotation is costly, though it is incremental as it builds on existing self-training methods.
The authors tackled the problem of training Handwritten Text Recognition models when labeled data is scarce by proposing a self-training approach using synthetic data and unlabeled samples, achieving results that significantly close the performance gap to fully-supervised models on four benchmark datasets.
Performances of Handwritten Text Recognition (HTR) models are largely determined by the availability of labeled and representative training samples. However, in many application scenarios labeled samples are scarce or costly to obtain. In this work, we propose a self-training approach to train a HTR model solely on synthetic samples and unlabeled data. The proposed training scheme uses an initial model trained on synthetic data to make predictions for the unlabeled target dataset. Starting from this initial model with rather poor performance, we show that a considerable adaptation is possible by training against the predicted pseudo-labels. Moreover, the investigated self-training strategy does not require any manually annotated training samples. We evaluate the proposed method on four widely used benchmark datasets and show its effectiveness on closing the gap to a model trained in a fully-supervised manner.