FaSTExt: Fast and Small Text Extractor

TL;DR

FaSTExt introduces a compact, efficient convolutional neural network for text detection in images, achieving high accuracy with significantly fewer parameters suitable for real-world applications.

Contribution

It presents a novel small-scale text extractor utilizing depthwise separable convolutions and multi-scale outputs, improving efficiency over larger models.

Findings

Effective text detection with 1.58 to 10.59 million parameters.

Maintains high accuracy while reducing model complexity.

Suitable for real-world applications due to efficiency.

Abstract

Text detection in natural images is a challenging but necessary task for many applications. Existing approaches utilize large deep convolutional neural networks making it difficult to use them in real-world tasks. We propose a small yet relatively precise text extraction method. The basic component of it is a convolutional neural network which works in a fully-convolutional manner and produces results at multiple scales. Each scale output predicts whether a pixel is a part of some word, its geometry, and its relation to neighbors at the same scale and between scales. The key factor of reducing the complexity of the model was the utilization of depthwise separable convolution, linear bottlenecks, and inverted residuals. Experiments on public datasets show that the proposed network can effectively detect text while keeping the number of parameters in the range of 1.58 to 10.59 million in different configurations.