Fused Text Segmentation Networks for Multi-oriented Scene Text Detection

Dai, Yuchen; Huang, Zheng; Gao, Yuting; Xu, Youxuan; Chen, Kai; Guo, Jie; Qiu, Weidong

doi:10.1109/icpr.2018.8546066

Cited by 121 publications

(55 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ExtraData R P F Proposed × 0.815 0.908 0.859 IncepText [37] × 0.806 0.905 0.853 FTSN [5] D 0.800 0.886 0.841 R2CNN [16] D 0.797 0.856 0.825 DDR [13] D 0.800 0.820 0.810 EAST [41] -0.783 0.832 0.807 RRPN [28] D 0.732 0.822 0.774 SegLink [33] D 0.731 0.768 0.749 Table 5: Comparison with prior arts on ICDAR-2015. R, P and F stand for recall, precision and F-measure respectively.…”

Section: Methodsmentioning

confidence: 99%

“…Deng et al [6] proposed to detect text by linking pixels within the same text instances together. Dai et al [5] and Yang et al [37] adopted the FCIS framework [19] to solve the text detection problem. In this paper, we borrowed Mask R-CNN, which is the latest state-of-the-art instance segmentation approach, to further enhance the text detection performance.…”

Section: Text Detectionmentioning

confidence: 99%

“…Although these approaches are composed of simpler pipelines, they still struggle with curved text detection. To solve this problem, some recent approaches like PixelLink [6], FTSN [5], and IncepText [37], propose to formulate text detection as an instance segmentation problem so that both straight text and curved text can be detected in a unified manner. Specifically, PixelLink proposes to detect text by linking pixels within the same text instances together, while FTSN and IncepText borrow the FCIS framework [19] to solve the text detection problem.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Mask R-CNN With Pyramid Attention Network for Scene Text Detection

Huang

Zhong

Sun

et al. 2019

2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

In this paper, we present a new Mask R-CNN based text detection approach which can robustly detect multioriented and curved text from natural scene images in a unified manner. To enhance the feature representation ability of Mask R-CNN for text detection tasks, we propose to use the Pyramid Attention Network (PAN) as a new backbone network of Mask R-CNN. Experiments demonstrate that PAN can suppress false alarms caused by text-like backgrounds more effectively. Our proposed approach has achieved superior performance on both multi-oriented (ICDAR-2015, ICDAR-2017 MLT) and curved (SCUT-CTW1500) text detection benchmark tasks by only using single-scale and single-model testing.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Text Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Mask R-CNN With Pyramid Attention Network for Scene Text Detection

Huang

Zhong

Sun

et al. 2019

2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

show abstract

“…F-measure versus training round for different strategies on Total-Text test set. SOTA here corresponds to [9] .…”

Section: B Implementation Detailsmentioning

confidence: 99%

Curved Text Detection in Natural Scene Images with Semi- and Weakly-Supervised Learning

Qin

Zhou

Yang

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

Detecting curved text in the wild is very challenging. Recently, most state-of-the-art methods are segmentation based and require pixel-level annotations. We propose a novel scheme to train an accurate text detector using only a small amount of pixel-level annotated data and a large amount of data annotated with rectangles or even unlabeled data. A baseline model is first obtained by training with the pixellevel annotated data and then used to annotate unlabeled or weakly labeled data. A novel strategy which utilizes groundtruth bounding boxes to generate pseudo mask annotations is proposed in weakly-supervised learning. Experimental results on CTW1500 and Total-Text demonstrate that our method can substantially reduce the requirement of pixel-level annotated data. Our method can also generalize well across two datasets. The performance of the proposed method is comparable with the state-of-the-art methods with only 10% pixel-level annotated data and 90% rectangle-level weakly annotated data.

show abstract

“…Some methods try to design a simple label calculation rule for oriented objects. For example, [6] adopts Mask R-CNN [7] for detecting oriented text lines. [8] regresses the outline of objects with multiple points on sliding lines.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Period Embedding for Representing Oriented Objects in Aerial Images

Zhu

2020

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

We propose a novel method for representing oriented objects in aerial images named Adaptive Period Embedding (APE). While traditional object detection methods represent object with horizontal bounding boxes, the objects in aerial images are oritented. Calculating the angle of object is an yet challenging task. While almost all previous object detectors for aerial images directly regress the angle of objects, they use complex rules to calculate the angle, and their performance is limited by the rule design. In contrast, our method is based on the angular periodicity of oriented objects. The angle is represented by two two-dimensional periodic vectors whose periods are different, the vector is continuous as shape changes. The label generation rule is more simple and reasonable compared with previous methods. The proposed method is general and can be applied to other oriented detector. Besides, we propose a novel IoU calculation method for long objects named length independent IoU (LIIoU). We intercept part of the long side of the target box to get the maximum IoU between the proposed box and the intercepted target box. Thereby, some long boxes will have corresponding positive samples. Our method reaches the 1 st place of DOAI2019 competition task1 (oriented object) held in workshop on Detecting Objects in Aerial Images in conjunction with IEEE CVPR 2019.

show abstract

Fused Text Segmentation Networks for Multi-oriented Scene Text Detection

Cited by 121 publications

References 46 publications

Mask R-CNN With Pyramid Attention Network for Scene Text Detection

Mask R-CNN With Pyramid Attention Network for Scene Text Detection

Curved Text Detection in Natural Scene Images with Semi- and Weakly-Supervised Learning

Adaptive Period Embedding for Representing Oriented Objects in Aerial Images

Contact Info

Product

Resources

About