Single Shot Text Detector with Regional Attention

By Pan He, Weilin Huang, Tong He, Qile Zhu, Yu Qiao, Xiaolin Li. 2017

315 citations recorded by the source corpus.

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, allowing 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 benchmark, advancing the state-of-the-art results in [18, 28]. Demo is available at: http://sstd.whuang.org/.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Adversarial Examples: Attacks and Defenses for Deep Learning (2017, background): “CNNs have shown incredible successes in computer vision tasks, such as image classification [1], [39], object detection [4], [40], text recognition [41], [42], and semantic segmentation [43], [44].”
Object Detection in 20 Years: A Survey (2019, methodology): “The first group of methods frame the text detection as a special case of general object detection [251, 348–357].”
Sliding Line Point Regression for Shape Robust Scene Text Detection (2018, methodology): “[25] used the attention map to remove background noise.”
Attention mechanisms in computer vision: A survey (2021, background): “, image classification [5, 6], object detection [7, 8], semantic segmentation [9, 10], face recognition [11, 12], person re-identification [13, 14], action recognition [15, 16], few-shot learning [17, 18], medical image processing [19, 20], image generation [21, 22], pose estimation [23], super reso”
Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes (2018, background): “A large number of methods have been proposed to detect scene text [7, 36, 37, 50, 19, 23, 54, 21, 47, 54, 56, 30, 52, 55, 34, 15, 48, 43, 57, 16, 35, 31].”
Real-Time Scene Text Detection With Differentiable Binarization and Adaptive Scale Fusion (2022, background): “Although the shrunk text regions are helpful to the cases that the text instance is not in the center region of another text instance, it fails when the TABLE 7 Detection Results on the ICDAR 2015 Dataset Method P R F FPS CTPN [45] 74.2 51.6 60.9 7.1 EAST [65] 83.6 73.5 78.2 13.2 SSTD [15] 80.2 73.9”
Efficient and Accurate Arbitrary-Shaped Text Detection With Pixel Aggregation Network (2019, background): “SSTD [16] generates text attention map to enhance the text region of the feature map and suppress background information, which is beneficial for tiny texts.”
Real-time Scene Text Detection with Differentiable Binarization (2019, background): “SSTD (He et al. 2017a) proposed an attention mechanism to roughly identi-ﬁes text regions.”
Scene Text Detection and Recognition: The Deep Learning Era (2018, methodology): “Contrarily, sub-text level detection methods [89], [20]9, [41], [148], [152], [44]10, [40], [121], [166], [133]11, [140], [171] only predicts parts that are combined to make a text instance.”
Shape Robust Text Detection With Progressive Scale Expansion Network (2018, methodology): “For the regression-based approaches [36, 43, 32, 16, 42, 23, 11, 13, 27], the text targets are usually represented in the forms of rectangles or quadrangles with certain orientations.”
Look More Than Once: An Accurate Detector for Text of Arbitrary Shapes (2019, unknown): “Recall Precision Hmean SegLink [30] 76.5 74.7 75.6 MCN [19] 80.0 72.0 76.0 SSTD [7] 73.9 80.2 76.9 WordSup [10] 77.0 79.3 78.2 EAST [42] 78.3 83.3 80.7 He et al. [8] 80.0 82.0 81.0 TextSnake [21] 80.4 84.9 82.6 PixelLink [3] 82.0 85.5 83.7 RRD [15] 80.0 88.0 83.8 Lyu et al. [23] 79.7 89.5 84.3…”
An End-to-End TextSpotter with Explicit Alignment and Attention (2018, background): “[8] proposed a text attention module capable of automatically learning the attention of rough text regions from the convolutional features.”
FOTS: Fast Oriented Text Spotting with a Unified Network (2018, unknown): “49 - 76 SSTD [13] 87 88 FCRNall+multi-filt [10] - - - - - 84.”
Character Region Awareness for Text Detection (2019, methodology): “Recently, scene text detectors based on deep learning have shown promising performance [8, 40, 21, 4, 11, 10, 12, 13, 17, 24, 25, 32, 26].”
TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes (2018, unknown): “52 SSTD [45] 80.”
Small Sample Learning in Big Data Era (2018, methodology): “Besides, He et al. (2017b) presented a fast yet accurate text detector with attention mechanism encoding strong supervised information of text in training that predicted wordlevel bounding boxes in one shot.”