SAFL: A Self-Attention Scene Text Recognizer with Focal Loss

TL;DR

SAFL introduces a self-attention neural network with focal loss and spatial transformer networks to improve scene text recognition, effectively handling distortions and irregular layouts with superior performance.

Contribution

The paper presents a novel self-attention based scene text recognizer with focal loss and spatial transformer networks, outperforming existing methods.

Findings

Achieves state-of-the-art performance on seven benchmarks.

Effectively handles distortions and irregular text layouts.

Focal loss improves training on low-frequency samples.

Abstract

In the last decades, scene text recognition has gained worldwide attention from both the academic community and actual users due to its importance in a wide range of applications. Despite achievements in optical character recognition, scene text recognition remains challenging due to inherent problems such as distortions or irregular layout. Most of the existing approaches mainly leverage recurrence or convolution-based neural networks. However, while recurrent neural networks (RNNs) usually suffer from slow training speed due to sequential computation and encounter problems as vanishing gradient or bottleneck, CNN endures a trade-off between complexity and performance. In this paper, we introduce SAFL, a self-attention-based neural network model with the focal loss for scene text recognition, to overcome the limitation of the existing approaches. The use of focal loss instead of negative log-likelihood helps the model focus more on low-frequency samples training. Moreover, to deal with the distortions and irregular texts, we exploit Spatial TransformerNetwork (STN) to rectify text before passing to the recognition network. We perform experiments to compare the performance of the proposed model with seven benchmarks. The numerical results show that our model achieves the best performance.