Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00024
Julien Maître, M. Ménard, Guillaume Chiron, A. Bouju, Nicolas Sidère
This paper is related to a project aiming at discovering weak signals from different streams of information, possibly sent by whistleblowers. The study presented in this paper tackles the particular problem of clustering topics at multi-levels from multiple documents, and then extracting meaningful descriptors, such as weighted lists of words for document representations in a multi-dimensions space. In this context, we present a novel idea which combines Latent Dirichlet Allocation and Word2vec (providing a consistency metric regarding the partitioned topics) as potential method for limiting the "a priori" number of cluster K usually needed in classical partitioning approaches. We proposed 2 implementations of this idea, respectively able to: (1) finding the best K for LDA in terms of topic consistency; (2) gathering the optimal clusters from different levels of clustering. We also proposed a non-traditional visualization approach based on a multi-agents system which combines both dimension reduction and interactivity.
{"title":"A Meaningful Information Extraction System for Interactive Analysis of Documents","authors":"Julien Maître, M. Ménard, Guillaume Chiron, A. Bouju, Nicolas Sidère","doi":"10.1109/ICDAR.2019.00024","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00024","url":null,"abstract":"This paper is related to a project aiming at discovering weak signals from different streams of information, possibly sent by whistleblowers. The study presented in this paper tackles the particular problem of clustering topics at multi-levels from multiple documents, and then extracting meaningful descriptors, such as weighted lists of words for document representations in a multi-dimensions space. In this context, we present a novel idea which combines Latent Dirichlet Allocation and Word2vec (providing a consistency metric regarding the partitioned topics) as potential method for limiting the \"a priori\" number of cluster K usually needed in classical partitioning approaches. We proposed 2 implementations of this idea, respectively able to: (1) finding the best K for LDA in terms of topic consistency; (2) gathering the optimal clusters from different levels of clustering. We also proposed a non-traditional visualization approach based on a multi-agents system which combines both dimension reduction and interactivity.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133812088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/icdar.2019.00025
Najah-Imane Bentabet, Rémi Juge, Sira Ferradans
The generation of precise and detailed Table-Of-Contents (TOC) from a document is a problem of major importance for document understanding and information extraction. Despite its importance, it is still a challenging task, especially for non-standardized documents with rich layout information such as commercial documents. In this paper, we present a new neural-based pipeline for TOC generation applicable to any searchable document. Unlike previous methods, we do not use semantic labeling nor assume the presence of parsable TOC pages in the document. Moreover, we analyze the influence of using external knowledge encoded as a template. We empirically show that this approach is only useful in a very low resource environment. Finally, we propose a new domain-specific data set that sheds some light on the difficulties of TOC generation in real-world documents. The proposed method shows better performance than the state-of-the-art on a public data set and on the newly released data set.
{"title":"Table-of-Contents Generation on Contemporary Documents","authors":"Najah-Imane Bentabet, Rémi Juge, Sira Ferradans","doi":"10.1109/icdar.2019.00025","DOIUrl":"https://doi.org/10.1109/icdar.2019.00025","url":null,"abstract":"The generation of precise and detailed Table-Of-Contents (TOC) from a document is a problem of major importance for document understanding and information extraction. Despite its importance, it is still a challenging task, especially for non-standardized documents with rich layout information such as commercial documents. In this paper, we present a new neural-based pipeline for TOC generation applicable to any searchable document. Unlike previous methods, we do not use semantic labeling nor assume the presence of parsable TOC pages in the document. Moreover, we analyze the influence of using external knowledge encoded as a template. We empirically show that this approach is only useful in a very low resource environment. Finally, we propose a new domain-specific data set that sheds some light on the difficulties of TOC generation in real-world documents. The proposed method shows better performance than the state-of-the-art on a public data set and on the newly released data set.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"215 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132803367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00206
Elvis Koci, Maik Thiele, Oscar Romero, Wolfgang Lehner
Spreadsheets are very successful content generation tools, used in almost every enterprise to create a wealth of information. However, this information is often intermingled with various formatting, layout, and textual metadata, making it hard to identify and interpret the tabular payload. Previous works proposed to solve this problem by mainly using heuristics. Although fast to implement, these approaches fail to capture the high variability of user-generated spreadsheet tables. Therefore, in this paper, we propose a supervised approach that is able to adapt to arbitrary spreadsheet datasets. We use a graph model to represent the contents of a sheet, which carries layout and spatial features. Subsequently, we apply genetic-based approaches for graph partitioning, to recognize the parts of the graph corresponding to tables in the sheet. The search for tables is guided by an objective function, which is tuned to match the specific characteristics of a given dataset. We present the feasibility of this approach with an experimental evaluation, on a large, real-world spreadsheet corpus.
{"title":"A Genetic-Based Search for Adaptive Table Recognition in Spreadsheets","authors":"Elvis Koci, Maik Thiele, Oscar Romero, Wolfgang Lehner","doi":"10.1109/ICDAR.2019.00206","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00206","url":null,"abstract":"Spreadsheets are very successful content generation tools, used in almost every enterprise to create a wealth of information. However, this information is often intermingled with various formatting, layout, and textual metadata, making it hard to identify and interpret the tabular payload. Previous works proposed to solve this problem by mainly using heuristics. Although fast to implement, these approaches fail to capture the high variability of user-generated spreadsheet tables. Therefore, in this paper, we propose a supervised approach that is able to adapt to arbitrary spreadsheet datasets. We use a graph model to represent the contents of a sheet, which carries layout and spatial features. Subsequently, we apply genetic-based approaches for graph partitioning, to recognize the parts of the graph corresponding to tables in the sheet. The search for tables is guided by an objective function, which is tuned to match the specific characteristics of a given dataset. We present the feasibility of this approach with an experimental evaluation, on a large, real-world spreadsheet corpus.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115633820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00044
M. Kassis, Jihad El-Sana
We present a fully automated learning free method, for line detection in manuscripts. We begin by separating components that span over multiple lines, then we remove noise, and small connected components such as diacritics. We apply a distance transform on the image to create the image skeleton. The skeleton is pruned, its vertexes and edges are detected, in order to generate the initial document graph. We calculate the vertex v-score using its t-score and l-score quantifying its distance from being an absolute link in a line. In a greedy manner we classify each edge in the graph either a link, a bridge or a conflict edge. We merge every two edges classified as link together, then we merge the conflict edges next. Finally we remove the bridge edges from the graph generating the final form of the graph. Each edge in the graph equals to one extracted line. We applied the method on the DIVA-hisDB dataset on both public and private sections. The public section participated in the recently conducted Layout Analysis for Challenging Medieval Manuscripts Competition, and we have achieved results surpassing the vast majority of these systems.
{"title":"Learning Free Line Detection in Manuscripts using Distance Transform Graph","authors":"M. Kassis, Jihad El-Sana","doi":"10.1109/ICDAR.2019.00044","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00044","url":null,"abstract":"We present a fully automated learning free method, for line detection in manuscripts. We begin by separating components that span over multiple lines, then we remove noise, and small connected components such as diacritics. We apply a distance transform on the image to create the image skeleton. The skeleton is pruned, its vertexes and edges are detected, in order to generate the initial document graph. We calculate the vertex v-score using its t-score and l-score quantifying its distance from being an absolute link in a line. In a greedy manner we classify each edge in the graph either a link, a bridge or a conflict edge. We merge every two edges classified as link together, then we merge the conflict edges next. Finally we remove the bridge edges from the graph generating the final form of the graph. Each edge in the graph equals to one extracted line. We applied the method on the DIVA-hisDB dataset on both public and private sections. The public section participated in the recently conducted Layout Analysis for Challenging Medieval Manuscripts Competition, and we have achieved results surpassing the vast majority of these systems.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115092749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00115
Hao Kong, Dongqi Tang, Xi Meng, Tong Lu
Deep neural networks have shown their powerful ability in scene character recognition tasks; however, in real life applications, it is often hard to find a large amount of high-quality scene character images for training these networks. In this paper, we proposed a novel end-to-end network named Generative Adversarial Recognition Networks (GARN) for accurate natural scene character recognition in an end-to-end way. The proposed GARN consists of a generation part and a classification part. For the generation part, the purpose is to produce diverse realistic samples to help the classifier overcome the overfitting problem. While in the classification part, a multinomial classifier is trained along with the generator in the form of a game to achieve better character recognition performance. That is, the proposed GARN has the ability to augment scene character data by its generation part and recognize scene characters by its classification part. It is trained in an adversarial way to improve recognition performance. The experimental results on benchmark datasets and the comparisons with the state-of-the-art methods show the effectiveness of the proposed GARN in scene character recognition.
{"title":"GARN: A Novel Generative Adversarial Recognition Network for End-to-End Scene Character Recognition","authors":"Hao Kong, Dongqi Tang, Xi Meng, Tong Lu","doi":"10.1109/ICDAR.2019.00115","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00115","url":null,"abstract":"Deep neural networks have shown their powerful ability in scene character recognition tasks; however, in real life applications, it is often hard to find a large amount of high-quality scene character images for training these networks. In this paper, we proposed a novel end-to-end network named Generative Adversarial Recognition Networks (GARN) for accurate natural scene character recognition in an end-to-end way. The proposed GARN consists of a generation part and a classification part. For the generation part, the purpose is to produce diverse realistic samples to help the classifier overcome the overfitting problem. While in the classification part, a multinomial classifier is trained along with the generator in the form of a game to achieve better character recognition performance. That is, the proposed GARN has the ability to augment scene character data by its generation part and recognize scene characters by its classification part. It is trained in an adversarial way to improve recognition performance. The experimental results on benchmark datasets and the comparisons with the state-of-the-art methods show the effectiveness of the proposed GARN in scene character recognition.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116430745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00244
Zheng Huang, Kai Chen, Jianhua He, X. Bai, Dimosthenis Karatzas, Shijian Lu, C. V. Jawahar
The ICDAR 2019 Challenge on "Scanned receipts OCR and key information extraction" (SROIE) covers important aspects related to the automated analysis of scanned receipts. The SROIE tasks play a key role in many document analysis systems and hold significant commercial potential. Although a lot of work has been published over the years on administrative document analysis, the community has advanced relatively slowly, as most datasets have been kept private. One of the key contributions of SROIE to the document analysis community is to offer a first, standardized dataset of 1000 whole scanned receipt images and annotations, as well as an evaluation procedure for such tasks. The Challenge is structured around three tasks, namely Scanned Receipt Text Localization (Task 1), Scanned Receipt OCR (Task 2) and Key Information Extraction from Scanned Receipts (Task 3). The competition opened on 10th February, 2019 and closed on 5th May, 2019. We received 29, 24 and 18 valid submissions received for the three competition tasks, respectively. This report presents the competition datasets, define the tasks and the evaluation protocols, offer detailed submission statistics, as well as an analysis of the submitted performance. While the tasks of text localization and recognition seem to be relatively easy to tackle, it is interesting to observe the variety of ideas and approaches proposed for the information extraction task. According to the submissions' performance we believe there is still margin for improving information extraction performance, although the current dataset would have to grow substantially in following editions. Given the success of the SROIE competition evidenced by the wide interest generated and the healthy number of submissions from academic, research institutes and industry over different countries, we consider that the SROIE competition can evolve into a useful resource for the community, drawing further attention and promoting research and development efforts in this field.
{"title":"ICDAR2019 Competition on Scanned Receipt OCR and Information Extraction","authors":"Zheng Huang, Kai Chen, Jianhua He, X. Bai, Dimosthenis Karatzas, Shijian Lu, C. V. Jawahar","doi":"10.1109/ICDAR.2019.00244","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00244","url":null,"abstract":"The ICDAR 2019 Challenge on \"Scanned receipts OCR and key information extraction\" (SROIE) covers important aspects related to the automated analysis of scanned receipts. The SROIE tasks play a key role in many document analysis systems and hold significant commercial potential. Although a lot of work has been published over the years on administrative document analysis, the community has advanced relatively slowly, as most datasets have been kept private. One of the key contributions of SROIE to the document analysis community is to offer a first, standardized dataset of 1000 whole scanned receipt images and annotations, as well as an evaluation procedure for such tasks. The Challenge is structured around three tasks, namely Scanned Receipt Text Localization (Task 1), Scanned Receipt OCR (Task 2) and Key Information Extraction from Scanned Receipts (Task 3). The competition opened on 10th February, 2019 and closed on 5th May, 2019. We received 29, 24 and 18 valid submissions received for the three competition tasks, respectively. This report presents the competition datasets, define the tasks and the evaluation protocols, offer detailed submission statistics, as well as an analysis of the submitted performance. While the tasks of text localization and recognition seem to be relatively easy to tackle, it is interesting to observe the variety of ideas and approaches proposed for the information extraction task. According to the submissions' performance we believe there is still margin for improving information extraction performance, although the current dataset would have to grow substantially in following editions. Given the success of the SROIE competition evidenced by the wide interest generated and the healthy number of submissions from academic, research institutes and industry over different countries, we consider that the SROIE competition can evolve into a useful resource for the community, drawing further attention and promoting research and development efforts in this field.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117073203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chinese keyword spotting is a challenging task as there is no visual blank for Chinese words. Different from English words which are split naturally by visual blanks, Chinese words are generally split only by semantic information. In this paper, we propose a new Chinese keyword spotter for natural images, which is inspired by Mask R-CNN. We propose to predict the keyword masks guided by text line detection. Firstly, proposals of text lines are generated by Faster R-CNN; Then, text line masks and keyword masks are predicted by segmentation in the proposals. In this way, the text lines and keywords are predicted in parallel. We create two Chinese keyword datasets based on RCTW-17 and ICPR MTWI2018 to verify the effectiveness of our method.
{"title":"A Multi-oriented Chinese Keyword Spotter Guided by Text Line Detection","authors":"Pei Xu, Shan Huang, Hongzhen Wang, Hao Song, Shen Huang, Qi Ju","doi":"10.1109/icdar.2019.00112","DOIUrl":"https://doi.org/10.1109/icdar.2019.00112","url":null,"abstract":"Chinese keyword spotting is a challenging task as there is no visual blank for Chinese words. Different from English words which are split naturally by visual blanks, Chinese words are generally split only by semantic information. In this paper, we propose a new Chinese keyword spotter for natural images, which is inspired by Mask R-CNN. We propose to predict the keyword masks guided by text line detection. Firstly, proposals of text lines are generated by Faster R-CNN; Then, text line masks and keyword masks are predicted by segmentation in the proposals. In this way, the text lines and keywords are predicted in parallel. We create two Chinese keyword datasets based on RCTW-17 and ICPR MTWI2018 to verify the effectiveness of our method.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115228273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
How to efficiently judge whether a natural image contains texts or not is an important problem. Since text detection and recognition algorithms are usually time-consuming, and it is unnecessary to run them on images that do not contain any texts. In this paper, we investigate this problem from two perspectives: the speed and the accuracy. First, to achieve high speed for efficient filtering large number of images especially on CPU, we propose using small and shallow convolutional neural network, where the features from different layers are adaptively pooled into certain sizes to overcome difficulties caused by multiple scales and various locations. Although this can achieve high speed but its accuracy is not satisfactory due to limited capacity of small network. Therefore, our second contribution is using the knowledge distillation to improve the accuracy of the small network, by constructing a larger and deeper neural network as teacher network to instruct the learning process of the small network. With the above two strategies, we can achieve both high speed and high accuracy for filtering scene text images. Experimental results on a benchmark dataset have shown the effectiveness of our method: the teacher network yields state-of-the-art performance, and the distilled small network achieves high performance while maintaining high speed which is 176 times faster on CPU and 3.8 times faster on GPU than a compared benchmark method.
{"title":"Fast Text/non-Text Image Classification with Knowledge Distillation","authors":"Miao Zhao, Rui-Qi Wang, Fei Yin, Xu-Yao Zhang, Lin-Lin Huang, J. Ogier","doi":"10.1109/ICDAR.2019.00234","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00234","url":null,"abstract":"How to efficiently judge whether a natural image contains texts or not is an important problem. Since text detection and recognition algorithms are usually time-consuming, and it is unnecessary to run them on images that do not contain any texts. In this paper, we investigate this problem from two perspectives: the speed and the accuracy. First, to achieve high speed for efficient filtering large number of images especially on CPU, we propose using small and shallow convolutional neural network, where the features from different layers are adaptively pooled into certain sizes to overcome difficulties caused by multiple scales and various locations. Although this can achieve high speed but its accuracy is not satisfactory due to limited capacity of small network. Therefore, our second contribution is using the knowledge distillation to improve the accuracy of the small network, by constructing a larger and deeper neural network as teacher network to instruct the learning process of the small network. With the above two strategies, we can achieve both high speed and high accuracy for filtering scene text images. Experimental results on a benchmark dataset have shown the effectiveness of our method: the teacher network yields state-of-the-art performance, and the distilled small network achieves high performance while maintaining high speed which is 176 times faster on CPU and 3.8 times faster on GPU than a compared benchmark method.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123514711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00222
Tonghua Su, Wei Pan, Lijuan Yu
Current state of handwritten Chinese character recognition (HCCR) conducted on well-confined character set, far from meeting industrial requirements. The paper describes the creation of a large-scale handwritten Chinese character database. Constructing the database is an effort to scale up Chinese handwritten character classification task to cover the full list of GBK character set specification. It consists of 21-thousand Chinese character categories and 20-million character images, larger than previous databases both in scale and diversity. We present solutions to the challenges of collecting and annotating such large-scale handwritten character samples. We elaborately design the sampling strategy, extract salient signals in a systematic way, annotate the tremendous characters through three distinct stages. Experiments are conducted the generalization to other handwritten character databases and our database demonstrates great values. Surely, its scale opens unprecedented opportunities both in evaluation of character recognition algorithms and in developing new techniques.
{"title":"HITHCD-2018: Handwritten Chinese Character Database of 21K-Category","authors":"Tonghua Su, Wei Pan, Lijuan Yu","doi":"10.1109/ICDAR.2019.00222","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00222","url":null,"abstract":"Current state of handwritten Chinese character recognition (HCCR) conducted on well-confined character set, far from meeting industrial requirements. The paper describes the creation of a large-scale handwritten Chinese character database. Constructing the database is an effort to scale up Chinese handwritten character classification task to cover the full list of GBK character set specification. It consists of 21-thousand Chinese character categories and 20-million character images, larger than previous databases both in scale and diversity. We present solutions to the challenges of collecting and annotating such large-scale handwritten character samples. We elaborately design the sampling strategy, extract salient signals in a systematic way, annotate the tremendous characters through three distinct stages. Experiments are conducted the generalization to other handwritten character databases and our database demonstrates great values. Surely, its scale opens unprecedented opportunities both in evaluation of character recognition algorithms and in developing new techniques.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130339133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.1109/ICDAR.2019.00235
Yuhuan Xiu, Qingqing Wang, Hongjian Zhan, Man Lan, Yue Lu
Handwritten Chinese text recognition (HCTR) has received extensive attention from the community of pattern recognition in the past decades. Most existing deep learning methods consist of two stages, i.e., training a text recognition network on the base of visual information, followed by incorporating language constrains with various language models. Therefore, the inherent linguistic semantic information is often neglected when designing the recognition network. To tackle this problem, in this work, we propose a novel multi-level multimodal fusion network and properly embed it into an attention-based LSTM so that both the visual information and the linguistic semantic information can be fully leveraged when predicting sequential outputs from the feature vectors. Experimental results on the ICDAR-2013 competition dataset demonstrate a comparable result with the state-of-the-art approaches.
{"title":"A Handwritten Chinese Text Recognizer Applying Multi-level Multimodal Fusion Network","authors":"Yuhuan Xiu, Qingqing Wang, Hongjian Zhan, Man Lan, Yue Lu","doi":"10.1109/ICDAR.2019.00235","DOIUrl":"https://doi.org/10.1109/ICDAR.2019.00235","url":null,"abstract":"Handwritten Chinese text recognition (HCTR) has received extensive attention from the community of pattern recognition in the past decades. Most existing deep learning methods consist of two stages, i.e., training a text recognition network on the base of visual information, followed by incorporating language constrains with various language models. Therefore, the inherent linguistic semantic information is often neglected when designing the recognition network. To tackle this problem, in this work, we propose a novel multi-level multimodal fusion network and properly embed it into an attention-based LSTM so that both the visual information and the linguistic semantic information can be fully leveraged when predicting sequential outputs from the feature vectors. Experimental results on the ICDAR-2013 competition dataset demonstrate a comparable result with the state-of-the-art approaches.","PeriodicalId":325437,"journal":{"name":"2019 International Conference on Document Analysis and Recognition (ICDAR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129747217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：