Scene Text Recognition with Sliding Convolutional Character Models

Abstract

Scene text recognition has attracted great interests from the computer vision and pattern recognition community in recent years. State-of-the-art methods use concolutional neural networks (CNNs), recurrent neural networks with long short-term memory (RNN-LSTM) or the combination of them. In this paper, we investigate the intrinsic characteristics of text recognition, and inspired by human cognition mechanisms in reading texts, we propose a scene text recognition method with character models on convolutional feature map. The method simultaneously detects and recognizes characters by sliding the text line image with character models, which are learned end-to-end on text line images labeled with text transcripts. The character classifier outputs on the sliding windows are normalized and decoded with Connectionist Temporal Classification (CTC) based algorithm. Compared to previous methods, our method has a number of appealing properties: (1) It avoids the difficulty of character segmentation which hinders the performance of segmentation-based recognition methods; (2) The model can be trained simply and efficiently because it avoids gradient vanishing/exploding in training RNN-LSTM based models; (3) It bases on character models trained free of lexicon, and can recognize unknown words. (4) The recognition process is highly parallel and enables fast recognition. Our experiments on several challenging English and Chinese benchmarks, including the IIIT-5K, SVT, ICDAR03/13 and TRW15 datasets, demonstrate that the proposed method yields superior or comparable performance to state-of-the-art methods while the model size is relatively small.

10 pages,4 figures

References in corpus (5)

• Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
• End-to-End Text Recognition with Hybrid HMM Maxout Models
• Recursive Recurrent Nets with Attention Modeling for OCR in the Wild
• Reading Scene Text in Deep Convolutional Sequences
• ICDAR 2015 Text Reading in the Wild Competition

Cited by in corpus (14)

• Rosetta: Large scale system for text detection and recognition in images
• Scene Text Detection and Recognition: The Deep Learning Era
• Towards Accurate Scene Text Recognition with Semantic Reasoning Networks
• Text Recognition in the Wild: A Survey
• Text Detection and Recognition in the Wild: A Review
• NRTR: A No-Recurrence Sequence-to-Sequence Model For Scene Text Recognition
• SCAN: Sliding Convolutional Attention Network for Scene Text Recognition
• A Multi-Object Rectified Attention Network for Scene Text Recognition
• Rethinking Irregular Scene Text Recognition
• From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network
• Gaussian Constrained Attention Network for Scene Text Recognition
• ShopSign: a Diverse Scene Text Dataset of Chinese Shop Signs in Street Views
• A New Perspective for Flexible Feature Gathering in Scene Text Recognition Via Character Anchor Pooling
• Adaptive Embedding Gate for Attention-Based Scene Text Recognition