STRIDE : Scene Text Recognition In-Device

TL;DR

This paper presents a lightweight, real-time scene text recognition system optimized for on-device deployment, combining convolution attention modules and a novel orientation classifier to achieve high accuracy with minimal computational resources.

Contribution

The authors introduce a compact, efficient scene text recognition model with convolution attention modules and an orientation classifier, enabling real-time on-device OCR with high accuracy.

Findings

Model has only 0.88M parameters and runs in 2.44 ms per word.

Achieves 88.4% accuracy on ICDAR-13 dataset.

Surpasses on-device inference time and memory footprint of existing OCR systems.

Abstract

Optical Character Recognition (OCR) systems have been widely used in various applications for extracting semantic information from images. To give the user more control over their privacy, an on-device solution is needed. The current state-of-the-art models are too heavy and complex to be deployed on-device. We develop an efficient lightweight scene text recognition (STR) system, which has only 0.88M parameters and performs real-time text recognition. Attention modules tend to boost the accuracy of STR networks but are generally slow and not optimized for device inference. So, we propose the use of convolution attention modules to the text recognition networks, which aims to provide channel and spatial attention information to the LSTM module by adding very minimal computational cost. It boosts our word accuracy on ICDAR 13 dataset by almost 2\%. We also introduce a novel orientation classifier module, to support the simultaneous recognition of both horizontal and vertical text. The proposed model surpasses on-device metrics of inference time and memory footprint and achieves comparable accuracy when compared to the leading commercial and other open-source OCR engines. We deploy the system on-device with an inference speed of 2.44 ms per word on the Exynos 990 chipset device and achieve an accuracy of 88.4\% on ICDAR-13 dataset.