TextMountain: Accurate Scene Text Detection via Instance Segmentation
arXiv:1811.12786
Abstract
In this paper, we propose a novel scene text detection method named TextMountain. The key idea of TextMountain is making full use of border-center information. Different from previous works that treat center-border as a binary classification problem, we predict text center-border probability (TCBP) and text center-direction (TCD). The TCBP is just like a mountain whose top is text center and foot is text border. The mountaintop can separate text instances which cannot be easily achieved using semantic segmentation map and its rising direction can plan a road to top for each pixel on mountain foot at the group stage. The TCD helps TCBP learning better. Our label rules will not lead to the ambiguous problem with the transformation of angle, so the proposed method is robust to multi-oriented text and can also handle well with curved text. In inference stage, each pixel at the mountain foot needs to search the path to the mountaintop and this process can be efficiently completed in parallel, yielding the efficiency of our method compared with others. The experiments on MLT, ICDAR2015, RCTW-17 and SCUT-CTW1500 databases demonstrate that the proposed method achieves better or comparable performance in terms of both accuracy and efficiency. It is worth mentioning our method achieves an F-measure of 76.85% on MLT which outperforms the previous methods by a large margin. Code will be made available.
References in corpus (21)
- TextBoxes++: A Single-Shot Oriented Scene Text Detector
- R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection
- Feature Pyramid Networks for Object Detection
- TextBoxes: A Fast Text Detector with a Single Deep Neural Network
- Semantic Instance Segmentation with a Discriminative Loss Function
- Holistically-Nested Edge Detection
- Semantic Understanding of Scenes through the ADE20K Dataset
- Accurate Text Localization in Natural Image with Cascaded Convolutional Text Network
- Deep Matching Prior Network: Toward Tighter Multi-oriented Text Detection
- Deep Direct Regression for Multi-Oriented Scene Text Detection
- PixelLink: Detecting Scene Text via Instance Segmentation
- Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes
- Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
- Weakly Supervised Instance Segmentation using Class Peak Response
- Shape Robust Text Detection with Progressive Scale Expansion Network
- ICDAR2017 Competition on Reading Chinese Text in the Wild (RCTW-17)
- Sliding Line Point Regression for Shape Robust Scene Text Detection
- Fused Text Segmentation Networks for Multi-oriented Scene Text Detection
- Unified Perceptual Parsing for Scene Understanding
- Detecting Oriented Text in Natural Images by Linking Segments
- Semi-convolutional Operators for Instance Segmentation