IncepText: A New Inception-Text Module with Deformable PSROI Pooling for Multi-Oriented Scene Text Detection

Abstract

Incidental scene text detection, especially for multi-oriented text regions, is one of the most challenging tasks in many computer vision applications. Different from the common object detection task, scene text often suffers from a large variance of aspect ratio, scale, and orientation. To solve this problem, we propose a novel end-to-end scene text detector IncepText from an instance-aware segmentation perspective. We design a novel Inception-Text module and introduce deformable PSROI pooling to deal with multi-oriented text detection. Extensive experiments on ICDAR2015, RCTW-17, and MSRA-TD500 datasets demonstrate our method's superiority in terms of both effectiveness and efficiency. Our proposed method achieves 1st place result on ICDAR2015 challenge and the state-of-the-art performance on other datasets. Moreover, we have released our implementation as an OCR product which is available for public access.

arXiv commentAccepted by IJCAI 2018

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection
• Deformable Convolutional Networks
• Text-Attentional Convolutional Neural Networks for Scene Text Detection
• Scene Text Detection via Holistic, Multi-Channel Prediction
• EAST: An Efficient and Accurate Scene Text Detector
• Deep Direct Regression for Multi-Oriented Scene Text Detection
• Fully Convolutional Instance-aware Semantic Segmentation
• Fused Text Segmentation Networks for Multi-oriented Scene Text Detection
• Detecting Oriented Text in Natural Images by Linking Segments

Cited by in corpus (10)

• A Single-Shot Arbitrarily-Shaped Text Detector based on Context Attended Multi-Task Learning
• Pyramid Mask Text Detector
• Scene Text Detection with Supervised Pyramid Context Network
• Look More Than Once: An Accurate Detector for Text of Arbitrary Shapes
• Arbitrary Shape Scene Text Detection with Adaptive Text Region Representation
• Exploring the Capacity of an Orderless Box Discretization Network for Multi-orientation Scene Text Detection
• Location-Aware Feature Selection Text Detection Network
• TextTubes for Detecting Curved Text in the Wild
• RPM-Net: Robust Pixel-Level Matching Networks for Self-Supervised Video Object Segmentation
• Efficient Scene Text Detection with Textual Attention Tower