Aggregated Text Transformer for Scene Text Detection

TL;DR

This paper introduces the Aggregated Text Transformer (ATTR), a multi-scale self-attention based model that improves scene text detection by effectively representing texts of various sizes and shapes in natural images.

Contribution

The paper proposes a novel multi-scale aggregation strategy within a Transformer framework for robust scene text detection, handling curved and dense text regions.

Findings

Effective detection of curved texts

Improved accuracy on public datasets

Robust multi-scale text representation

Abstract

This paper explores the multi-scale aggregation strategy for scene text detection in natural images. We present the Aggregated Text TRansformer(ATTR), which is designed to represent texts in scene images with a multi-scale self-attention mechanism. Starting from the image pyramid with multiple resolutions, the features are first extracted at different scales with shared weight and then fed into an encoder-decoder architecture of Transformer. The multi-scale image representations are robust and contain rich information on text contents of various sizes. The text Transformer aggregates these features to learn the interaction across different scales and improve text representation. The proposed method detects scene texts by representing each text instance as an individual binary mask, which is tolerant of curve texts and regions with dense instances. Extensive experiments on public scene text detection datasets demonstrate the effectiveness of the proposed framework.