Single Shot Text Detector with Regional Attention
This work addresses the challenge of accurate text detection in images, particularly for small text, which is important for applications like document analysis and scene understanding, and it is incremental by improving upon existing FCN-based methods.
The paper tackles the problem of detecting text in natural images by introducing a single-shot text detector that directly outputs word-level bounding boxes, achieving an F-measure of 77% on the ICDAR 2015 benchmark and advancing state-of-the-art results.
We present a novel single-shot text detector that directly outputs word-level bounding boxes in a natural image. We propose an attention mechanism which roughly identifies text regions via an automatically learned attentional map. This substantially suppresses background interference in the convolutional features, which is the key to producing accurate inference of words, particularly at extremely small sizes. This results in a single model that essentially works in a coarse-to-fine manner. It departs from recent FCN- based text detectors which cascade multiple FCN models to achieve an accurate prediction. Furthermore, we develop a hierarchical inception module which efficiently aggregates multi-scale inception features. This enhances local details, and also encodes strong context information, allow- ing the detector to work reliably on multi-scale and multi- orientation text with single-scale images. Our text detector achieves an F-measure of 77% on the ICDAR 2015 bench- mark, advancing the state-of-the-art results in [18, 28]. Demo is available at: http://sstd.whuang.org/.