Home    People   Research   Publications   中文首页

 

   

Introduction:

    Our research mainly includes pattern recognition and computer vision, and information retrieval and data mining. We are specifically interested in building applicable and novel technologies and systems for related topics, and have been developing several state-of-the-art and applicable technologies, e.g., Scene Text Detection and Recognition, Financial Document Image Analysis and Recognition, Social Book Search, and Fast Face Recognition.

    At the moment, we focus on Intelligent Text Reading in ubiquitous multimedia (detection, recognition and retrieval and mining  in natural scenes, web images, ubiquitous documents and videos); Social Multimedia Retrieval for web books, products, and massive videos; and Online Face, Audio and Handwriting Verification for online security and social networks.

News:

  • 2017-04: Our paper is accepted by ACM SIGIR 2017. (Bo-Wen Zhang, Xu-Cheng Yin*, Fang Zhou, and Jianlin Jin, "Building your own reading list anytime via embedding relevance, quality, timeliness and diversity," ACM SIGIR 2017, accepted (short paper).)
  • 2017-04: Our two papers are accepted by IEEE TPAMI and IEEE TIP. IEEE TPAMI and IEEE TIP are top-tier journals in, Pattern Recognition, Computer Vision, Image Processing and related fields.  ([1] Shu Tian, Xu-Cheng Yin*, Ya Su, and Hong-Wei Hao, "A unified framework for tracking based text detection and recognition from web videos," IEEE TPAMI, accepted, 2017;  [2] Xu-Cheng Yin, Chun Yang, Wei-Yi Pei, Shu Tian, Ze-Yu Zuo, Chao Zhu and Junchi Yan, "Tracking based multi-orientation scene text detection: A unified framework with dynamic programming," IEEE TIP, accepted, 2017.)
  • 2016-11: Our paper, "Robust segmentation for video captions with complex backgrounds" (Zong-Heng Xing, Fang Zhou, Shu Tian and Xu-Cheng Yin) won the "Best Student Paper Award" of the 7th Conference on Pattern Recognition (CCPR'16).
  • 2016-10: Bo-Wen Zhang (a PhD candidate) in our lab receives "2016 Graduate National Scholarship", and Zan-Xia Jin (a master candidate) wins "2016 SanJin Scholarship" of our university.
  • 2016-10: Prof. Xu-Cheng Yin's team won several batches of 2016 BioASQ Challenge Task 3b: Biomedical Semantic QA. The BioQSQ Challenge is a challenge on large-scale biomedical semantic indexing and question answering.
  • 2016-07: Welcome Dr Shu TIAN to join our lab. Dr. TIAN received Ph.D. degree from University of Science and Technology Beijing, China in 2016.  His research mainly includes object tracking, pattern recognition, and multimedia understanding.
  • 2016-06: Welcome Dr. Chao ZHU to join our lab. Dr. ZHU received Ph.D. degree from École Centrale de Lille, France. He spent three years in Peking University, China as a post-doctor researcher before joining our university this year. His research mainly includes image processing and object detection.
  • 2016-05: Our team won 2016 CLEF Social Book Search (Suggestion Task), and all our 6 runs ranked top 6 in the evaluation.
  • 2016-04: One overview paper of Prof. Xu-Cheng Yin's team, "Text Detection, Tracking and Recognition in Video: A Comprehensive Survey", is accepted by IEEE Transactions on Image Processing (IEEE TIP), one top-tier journal in Image Processing, Pattern Recognition, and related fields. This survey presents a comprehensive survey of text detection, tracking and recognition from (scene and web) videos within a generic framework for uniformly describing detection, tracking, recognition and their relations and interactions.
  • 2016-04: One paper of Prof. Xu-Cheng Yin's team, "Scene Text Detection in Video by Learning Locally and Globally", is accepted by the 25th International Joint Conference on Artificial Intelligence (IJCAI'16). IJCAI is  a top-tier conference sponsored by AAAI and IJCAI in the areas of Artificial Intelligence and related fields.
  • 2016-03: Recently, a paper of Prof. Xu-Cheng Yin's team, "Robust Text Detection in Natural Scene Images" (IEEE TPAMI 2014), is recognized as a highly cited paper from ESI (Essential Science Indicators). This paper presents a novel robust text detection approach from scene images, and also won the first place of both "Text Localization in Real Scenes" and "Text Localization in Born-Digital Images (Web and Email)" in the ICDAR 2013 Robust Reading Competition. See details from the news of our university (in Chinese).
  • 2016-01: Prof. Xu-Cheng Yin, as a tutor representative, presents a talk on the Graduate Commencement of our university in 2016 Spring. The topic of his talk is, "Focusing, Insisting, and Breaking through". See details from the web page of our university (in Chinese).
  • 2015-12: Prof. Xu-Cheng Yin's team won several batches of 2015 BioASQ Challenge Task 3b: Biomedical Semantic QA. The BioQSQ Challenge is a challenge on large-scale biomedical semantic indexing and question answering.
  • 2015-12 (Dec 1): Dr. Weilin Huang from Shenzhen Institutes of Advance Technology, Chinese Academy of Sciences, presents a talk ("Scene Text Detection and Recognition with Deep Learing") in our lab. Dr. Weilin Huang is also a member of Multimedia Laboratory, The Chinese University of Hongkong, and his research interests are in computer vision and deep learning, including scene text detection and recognition, face recognition, object detection and image classification.
  • 2015-11 (Nov 17): Prof. Lianwen Jin from Department of Information Engineering, South China University of Technology, presents a talk ("Handwritten Chinese Character Recognition with Deep Learning and Path Signature Feature") in our lab. Prof. Lianwen Jin's research focuses on Online Character Recognition, Document Analysis and Recognition, and Deep Learning with DAR applications.
  • 2015-10: Wei-Yi Pei (a PhD candidate) and Ze-Yu Zuo (a master candidate) in our lab, both receive "2015 Graduate National Scholarship" of our university.
  • 2015-10: Faculty positions (Assistant/Associate/Full Professors) in Computer Vision and Machine Learning, or Information Retrieval and NLP available in our lab at the Department of Computer Science, University of Science and Technology Beijing. See details from http://prir.ustb.edu.cn/job/PrIrJob201510.htm (in Chinese).
  • 2015-08: On 2015 International Conference on Document Analysis and Recognition (ICDAR 2015), our technology (Prof. Yin's team) won the first place of both "End-To-End Focused Scene Text Recognition (Generic)" (Challenge 2) and "End-To-End Born-Digital Image Text Recognition (Generic)" (Challenge 1) in the ICDAR 2015 Robust Reading Competition. Our technology  also won the first place of "End-To-End Born-Digital Image Text Recognition (Weak)" and "Video Text Detection" in the ICDAR 2015 Robust Reading Competition.
  • 2015-08 (Aug 3): Prof. Hong YU from BioNLP Lab, University of Massachusetts Medical School, USA (and also an adjunct professor from School of Computer Science, University of Massachusetts Amherst), presents a talk ("Biomedical Natural Language Figure Processing Assisting High-Throughput Data Analysis") in our lab. Prof. Hong Yu's research focuses on information retrieval and natural language processing and their applications to the biomedical and healthcare domain.
  • 2015-07: We build a Social (Chinese and English) Book Search system (online testing demo), where a user can input a query with natural languages. Our system search relative books based on both book contents and social information (e.g., tags, reviews and scores).
  • 2015-06: Prof. Xu-Cheng Yin is awarded as "2015 The Best Favorite Advisor of Graduates" (ONE from TEN) of University of Science and Technology Beijing.
  • 2015-05: Our paper, "DeTEXT: A Database for Evaluating Text Extraction from Biomedical Literature Figures", cooperated with Prof. Erik Learned-Miller (Computer Vision Lab, School of Computer Science, University of Massachusetts Amherst, USA) and Prof. Hong YU (BioNLP Lab, University of Massachusetts Medical School, USA), is published in PLoS ONE, one famous open source journal.
  • 2015-01: Our paper, "Multi-orientation scene text detection with adaptive clustering", is accepted by IEEE Trans. Pattern Analysis and Machine Intelligence (TPAMI, 2013 SCI impact factor: 5.696), one top-tier journal in Artificial Intelligence, Pattern Recognition, Computer Vision and Machine Learning.
  • 2014-10: Chun Yang (a PhD candidate) and Bo-We Zhang (a master candidate) in our lab, both receive "2014 Graduate National Scholarship" of our university.
  • 2014-09: Dr. Xu-Cheng Yin has an invited online talk at VALSE for scene text detection and recognition. VALSE (Vision And Learning SEminar) is a famous (online/offline) platform for young Chinese scholars in the Computer Vision, Pattern Recognition and Machine Learning fields.
  • 2014-08: Our paper, "Social book reranking with generalized content-based filtering", is accepted (ORAL paper) by ACM International Conference on Information and Knowledge Management (CIKM'14). CIKM is a top-tier conference sponsored by ACM in the areas of Information Retrieval, Knowledge Management and Databases.
  • 2014-05: Dr Xu-Cheng Yin's team (with "Social Book Search and Recommendation Technology") won 2014 CLEF/INEX Social Book Search (Suggestion Task), and all our 6 runs ranked top 7 in the evaluation.
  • 2014-04: Dr. Xu-Cheng Yin's team is cooperated with and supported by iFlyTek Co., LTD. with Handwritten Document Analysis, Segmentation and Recognition technologies and applications.
  • 2014-03: Dr. Xu-Cheng Yin won Top 10 Newsmakers 2013 of our university (University of Science and Technology Beijing) as his team won the first place of "Text Localization in Real Scenes", "Text Localization in Born-Digital Images (Web and Email)", and "Text Segmentation in Born-Digital Images (Web and Email)" in the ICDAR 2013 Robust Reading Competition.
  • 2014-01: Dr. Xu-Cheng Yin's team is cooperated with and supported by Samsung Research China, with the "robust and applicable text detection engine in natural scenes" project for improving scene text understanding technologies and applications in smart phones and tablets.
  • 2013-12: Our paper, "Convex ensemble learning with sparsity and diversity", is accepted by Information Fusion (INFFUS, 2013 SCI impact factor: 3.472), one famous journal in Intelligent Information Processing, Data Mining and Pattern Recognition.
  • 2013-12: Xuwang YIN, a master candidate in our lab, receive "2013 Top 10 Graduate Academic Star" of our university. He is the first master candidate who received this honor in the "Top 10 Graduate Academic Star" history of our university.
  • 2013-10: Welcome Dr. Ya SU to join our lab. Dr. SU received Ph.D., M.Sc. and B.Sc. degrees all from Xi'dian University, China. He spent three years in Tsinghua University, China as a post-doctor researcher before joining our university this year. His research mainly includes face recognition and multimedia understanding.
  • 2013-09: Our paper, "Robust text detection in natural scene images", is accepted by IEEE Trans. Pattern Analysis and Machine Intelligence (TPAMI, 2013 SCI impact factor: 5.696), one top-tier journal in Artificial Intelligence, Pattern Recognition, Computer Vision and Machine Learning.
  • 2013-09: Several famous Chinese news medias (e.g., Science and Technology DailyChina Science Daily, and Xinhuanet.com) reported our scene text detection and segmentation technology ("USTB_TexStar" and "USTB_FuStar") which won the first place in ICDAR 2013 Robust Reading Competition.
  • 2013-08: On 2013 International Conference on Document Analysis and Recognition (ICDAR 2013), our technology ("USTB_TexStar") won the first place of both "Text Localization in Real Scenes" (Challenge 2) and "Text Localization in Born-Digital Images (Web and Email)" (Challenge 1) in the ICDAR 2013 Robust Reading Competition. Our technology ("USTB_FuStar") also won the first place of "Text Segmentation in Born-Digital Images (Web and Email)" (Challenge 1) in the ICDAR 2013 Robust Reading Competition.

©2013~2015 Pattern Recognition and Information Retrieval Lab, Department of Computer Science and Technology @ University of Science and Technology Beijing

Room 1005 Information Building, No. 30 Xueyuan Rd., Haidian Dist., Beijing 100083, China.