Sequential Visual and Semantic Consistency for Semi-supervised Text Recognition

Authors:
  • MYMingkun Yang
    Mingkun Yang
  • BYBiao Yang
    Biao Yang
  • MLMinghui Liao
    Minghui Liao
  • YZYingying Zhu
    Yingying Zhu
  • XBXiang Bai
    Xiang Bai

Abstract
Scene text recognition (STR) is a challenging task that requires large-scale annotated data for training. However, collecting and labeling real text images is expensive and time-consuming, which limits the availability of real data. Therefore, most existing STR methods resort to synthetic data, which may introduce domain discrepancy and degrade the performance of STR models. To alleviate this problem, recent semi-supervised STR methods exploit unlabeled real data by enforcing character-level consistency regularization between weakly and strongly augmented views of the same image. However, these methods neglect word-level consistency, which is crucial for sequence recognition tasks. This paper proposes a novel semi-supervised learning method for STR that incorporates word-level consistency regularization from both visual and semantic aspects. Specifically, we devise a shortest path alignment module to align the sequential visual features of different views and minimize their distance. Moreover, we adopt a reinforcement learning framework to optimize the semantic similarity of the predicted strings in the embedding space. We conduct extensive experiments on several standard and challenging STR benchmarks and demonstrate the superiority of our proposed method over existing semi-supervised STR methods.
All Related Versions

This publication has 22 references indexed in Scilit: