PixelLink: Detecting Scene Text via Instance Segmentation

By Dan Deng, Haifeng Liu, Xuelong Li, Deng Cai. AAAI Conference on Artificial Intelligence, 2018

598 citations recorded by the source corpus.

Most state-of-the-art scene text detection algorithms are deep learning based methods that depend on bounding box regression and perform at least two kinds of predictions: text/non-text classification and location regression. Regression plays a key role in the acquisition of bounding boxes in these methods, but it is not indispensable because text/non-text prediction can also be considered as a kind of semantic segmentation that contains full location information in itself. However, text instances in scene images often lie very close to each other, making them very difficult to separate via semantic segmentation. Therefore, instance segmentation is needed to address this problem. In this paper, PixelLink, a novel scene text detection algorithm based on instance segmentation, is proposed. Text instances are first segmented out by linking pixels within the same instance together. Text bounding boxes are then extracted directly from the segmentation result without location regression. Experiments show that, compared with regression-based methods, PixelLink can achieve better or comparable performance on several benchmarks, while requiring many fewer training iterations and less training data.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/a4a88145718ec8eff1228267bf3fe9f380b9495f

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

1. Chunhua Shen
2. Luc Van Gool
3. Junchi Yan
4. Dacheng Tao
5. Yu Qiao
6. Xue Yang
7. Xiang Bai
8. Alan L. Yuille

In Media Technology

1. Cong Yao
2. Xin He
3. Xue Yang
4. Shangbang Long
5. Junchi Yan
6. Chunhua Shen
7. Chenglin Liu
8. Tong Lu

Citations in the corpus

• Real-Time Scene Text Detection With Differentiable Binarization and Adaptive Scale Fusion (2022, unknown): “1 RRPN [40] 82 68 74 RRD [27] 87 73 79 10 MCN [33] 88 79 83 PixelLink [10] 83 73.”
• Look More Than Once: An Accurate Detector for Text of Arbitrary Shapes (2019, unknown): “…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 IncepText [37] 84.3 89.4 86.8 Mask TextSpotter [22] 81.0 91.6 86.0…”
• Efficient and Accurate Arbitrary-Shaped Text Detection With Pixel Aggregation Network (2019, methodology): “To separate adjacent text instances, PixelLink [3] performs text/non-text and links prediction in pixel level, then applies post-processing to obtain text boxes and excludes noises.”
• Arbitrary Shape Scene Text Detection With Adaptive Text Region Representation (2019, methodology): “Segmentation based methods [3, 19, 33] try to solve the problem by segmenting text regions from the background and an additional step is needed to get the final bounding boxes.”
• SCRDet: Towards More Robust Detection for Small, Cluttered and Rotated Objects (2018, background): “Another reason is that the objects in aerial images are often more closer to each other than in scene texts, which limits the applicability of segmentation based detection algorithm [7, 44] that otherwise work well on scene texts.”
• 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].”
• Real-time Scene Text Detection with Differentiable Binarization (2019, unknown): “s. Method P R F FPS (He et al. 2016b) 71 61 69 - DeepReg (He et al. 2017b) 77 70 74 1.1 RRPN (Ma et al. 2018) 82 68 74 - RRD (Liao et al. 2018) 87 73 79 10 MCN (Liu et al. 2018) 88 79 83 - PixelLink (Deng et al. 2018) 83 73.2 77.8 3 Corner (Lyu et al. 2018b) 87.6 76.2 81.5 5.7 TextSnake (Long et al.”
• Shape Robust Text Detection With Progressive Scale Expansion Network (2018, background): “0 PixelLink [4] - 82.”
• TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes (2018, methodology): “The models of EAST [47], SegLink [29], and PixelLink [2] are taken as baselines, since these two methods were also trained on ICDAR 2015.”