Stratified Domain Adaptation: A Progressive Self-Training Approach for Scene Text Recognition

TL;DR

This paper introduces Stratified Domain Adaptation, a progressive self-training method that partitions data based on domain proximity to improve scene text recognition across different domains.

Contribution

It proposes a novel stratification strategy and domain discriminator-based estimation to facilitate gradual domain adaptation in scene text recognition.

Findings

Significantly improves baseline scene text recognition models.

Effective in handling large domain gaps.

Outperforms existing unsupervised domain adaptation methods.

Abstract

Unsupervised domain adaptation (UDA) has become increasingly prevalent in scene text recognition (STR), especially where training and testing data reside in different domains. The efficacy of existing UDA approaches tends to degrade when there is a large gap between the source and target domains. To deal with this problem, gradually shifting or progressively learning to shift from domain to domain is the key issue. In this paper, we introduce the Stratified Domain Adaptation (StrDA) approach, which examines the gradual escalation of the domain gap for the learning process. The objective is to partition the training data into subsets so that the progressively self-trained model can adapt to gradual changes. We stratify the training data by evaluating the proximity of each data sample to both the source and target domains. We propose a novel method for employing domain discriminators to estimate the out-of-distribution and domain discriminative levels of data samples. Extensive experiments on benchmark scene-text datasets show that our approach significantly improves the performance of baseline (source-trained) STR models.