基本信息

周宇

博士生导师,副研究员

中国科学院信息工程研究所 第三研究室


研究方向为人工智能、计算机视觉、

深度学习与人工智能安全,专注于:

1)场景文字处理、提取与理解

2)视觉目标检测与视频理解等

3)自监督、增量与对抗学习等


电子邮箱:zhouyu@iie.ac.cn

DBLP & Google Scholar

招生信息

欢迎国科大考研学生提前进组科研实践!

-具备推免资格,满足国科大及信工所推免条件;

-具备自我激励能力,具备较好的逻辑、编程、数学、英语、语文相关能力;

-专业为计算机、人工智能、网络空间安全、自动化、电子等优先;

-直博优先,大四可以实习/毕业设计者优先;

-简历:含项目、竞赛、论文等信息,注明专业及排名(排名/总人数)、英语六/四级分数高考省份及省排名等信息。

个人履历

教育经历

  • 2009.12,哈尔滨工业大学,学士、硕士、博士

工作经历

  • 2012.03,上海交通大学,博士后
  • 2012.04至今,中国科学院信息工程研究所,助理研究员、副研究员、硕士生导师、博士生导师

学术服务

学术报告

  • 2022年04月24日,低质量场景文字识别技术研究”,中国图象图形学会“OCR学术前沿及产业应用”高峰论坛(报告视频),在线观众峰值8000余人

  • 2022年06月16日,“场景文字检测、识别与理解技术研究”,企业交流报告

专委会

  • 中国图象图形学会文档图像分析与识别专委会,委员

会议

  • Area Chair:ICME-22、ICME-21
  • SPC Member:IJCAI-21
  • PC Member:CVPR-22、ECCV-22、AAAI-22、IJCAI-22、ACM MM-22、CVPR-21、ICCV-21、AAAI-21、ACM MM-21、ICDAR-21、ChinaMM-21、ICPR-20...
  • Session Chair:ICME-21、TrustCom-14

期刊

  • Reviewer:IEEE TMM/TCSVT/TITS/MM、ACM TOMM、PR、IJPRAI ...
  • 审稿人:计算机学报、中国图象图形学报

学术论文

2022

  1. 周宇*,吕嘉昊,申化文,王威,魏谨,曾港艳,曾维超,王伟平. "从检测、识别到理解:场景文字相关领域研究进展." 中国自动化学会模式识别与机器智能专委会通讯特约专栏, 2022. (链接)
  2. W Wang, Y Zhou*, J Lv, D Wu, G Zhao, N Jiang, W Wang. "TPSNet: Reverse Thinking of Thin Plate Splines for Arbitrary Shape Scene Text Representation." ACM MM, 2022. (CCF-A)
  3. J Wei, Y Zhang, Y Zhou*, G Zeng, Z Qiao, Y Guo, H Wu, H Wang, W Wang. "TextBlock: Towards Scene Text Spotting without Fine-grained Detection." ACM MM, 2022. (CCF-A)
  4. B Fang, W Wu, C Liu, Y Zhou*, D He, W Wang. "MaMiCo: Macro-to-Micro Semantic Correspondence for Self-supervised Video Representation Learning." ACM MM, 2022. (CCF-A)
  5. X Chen, Y Zhou, D Wu, W Zhang, Y Zhou, B Li, W Wang. "Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification." AAAI, 2022. (CCF-A)
  6. D Yang, Y Zhou*, A Zhang, X Sun, D Wu, W Wang, Q Ye. "Multi-View Correlation Distillation for Incremental Object Detection." PR, 2022. (SCI一区, CCF-BPDF)
  7. Y Zhou, X Li, Y Zhou, Y Wang, Q Hu, W Wang. "Deep Collaborative Multi-Task Network: A Human Decision Process Inspired Model for Hierarchical Image Classification." PR, 2022. (SCI一区, CCF-BPDF)
  8. D Yang, Y Zhou*, W Shi, D Wu, W Wang. "RD-IOD: Two-Level Residual-Distillation-based Triple Network for Incremental Object Detection." TOMM, 2022. (SCI一区, CCF-B, PDF)
  9. D Luo, Y Zhou*, B Fang, Y Zhou, D Wu, W Wang. "Exploring Relations in Untrimmed Videos for Self-Supervised Learning." TOMM, 2022. (SCI一区, CCF-B, PDF)
  10. C Liu, Y Yao, D Luo, Y Zhou, Q Ye. "Self-supervised Motion Perception for Spatio-temporal Representation Learning." TNNLS, 2022. (SCI一区, CCF-B, PDF)
  11. Y Guo, Y Zhou*, X Qin, E Xie, W Wang. "UNITS: Unsupervised Intermediate Training Stage for Scene Text Detection." ICME, 2022. (CCF-B, Oral Presentation, PDF)
  12. C Fang, G Zeng, Y Zhou*, D Wu, C Ma, D Hu, W Wang."Towards Escaping from Language Bias and OCR Error: Semantics-Centered Text Visual Question Answering." ICME, 2022. (CCF-B, PDF)
  13. W Li, D Luo, B Fang, X Li, Y Zhou*, W Wang. "Video Motion Perception for Self-supervised Representation Learning." ICANN, 2022. (CCF-C, PDF)

2021

  1. Z Qiao, Y Zhou*, J Wei, W Wang, Y Zhang, N Jiang, H Wang, W Wang. "PIMNet: A Parallel, Iterative and Mimicking Network for Scene Text Recognition." ACM MM, 2021. (CCF-A, Best Paper Candidate [5/1942=2.5‰], PDF)
  2. G Zeng, Y Zhang, Y Zhou*, X Yang. "Beyond OCR + VQA: Involving OCR into the Flow for Robust and Accurate TextVQA." ACM MM, 2021. (CCF-A, Oral Presentation, Acceptance Rate 9.2%, PDF)
  3. X Li, Y Zhou*, Y Zhang, A Zhang, W Wang, N Jiang, H Wu, W Wang. "Dense Semantic Contrast for Self-Supervised Visual Representation Learning." ACM MM, 2021. (CCF-A, Oral Presentation, Acceptance Rate 9.2%, PDF)
  4. X Qin, Y Zhou*, Y Guo, D Wu, Z Tian, N Jiang, H Wang, W Wang. "Mask is All You Need: Rethinking Mask R-CNN for Dense and Arbitrary-Shaped Scene Text Detection." ACM MM, 2021. (CCF-A, PDF)
  5. W Zhang, D Wu, Y Zhou, B Li, W Wang, D Meng. "Binary Neural Network Hashing for Image Retrieval." SIGIR, 2021. (CCF-APDF)
  6. X Qin, Y Zhou*, Y Guo, D Wu, W Wang. "FC2RN: A Fully Convolutional Corner Refinement Network for Accurate Multi-Oriented Scene Text Detection." ICASSP, 2021. (CCF-BPDF)
  7. G Zeng, Y Zhang, Y Zhou*, X Yang. "A Cost-Efficient Framework for Scene Text Detection in the Wild." PRICAI, 2021. (CCF-C, PDF)
  8. Y Guo, Y Zhou*, X Qin, W Wang. "Which and Where to Focus: A Simple yet Accurate Framework for Arbitrary-Shaped Nearby Text Detection in Scene Images." ICANN, 2021. (CCF-C, PDF)
  9. X Li, Y Zhou, Y Zhou, W Wang. "MMF: Multi-Task Multi-Structure Fusion for Hierarchical Image Classification." ICANN, 2021. (CCF-C, PDF)
  10. H Li, Y Guo, Y Zhou*, W Wang. "Density-Net: A Density-Aware Network for 3D Object Detection." ICTAI, 2021. (CCF-C, PDF)
  11. Y Zhang, Y Zhou*, W Wang. "Exploring Instance Relations for Unsupervised Feature Embedding." arXiv preprint, 2021. (PDF)
2020
  1. Z Qiao, Y Zhou*, D Yang, Y Zhou, W Wang. "SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognition." CVPR, 2020. (CCF-A, Acceptance Rate 22%, 100 CitationsPDF)
  2. Y Yao, C Liu, D Luo, Y Zhou, Q Ye. "Video Playback Rate Perception for Self-Supervised Spatio-Temporal Representation Learning." CVPR, 2020. (CCF-A, Acceptance Rate 22%, 105 CitationsPDF)
  3. D Luo, C Liu, Y Zhou*, D Yang, C Ma, Q Ye, W Wang. "Video Cloze Procedure for Self-Supervised  Spatio-Temporal Learning." AAAI, 2020. (CCF-A, Oral Presentation, Acceptance Rate 5.8%, 110 CitationsPDF)
  4. W Zhang, D Wu, Y Zhou, B Li, W Wang, D Meng. "Deep Unsupervised Hybrid-similarity Hadamard Hashing." ACM MM, 2020. (CCF-APDF)
  5. S Zhao, D Wu, W Zhang, Y Zhou, B Li, W Wang. "Asymmetric Deep Hashing for Efficient Hash Code Compression." ACM MM, 2020. (CCF-A, PDF)
  6. Y Chen, W Wang, Y Zhou*, F Yang, D Yang, W Wang. "Self-Training for Domain Adaptive Scene Text Detection." ICPR, 2020. (CCF-C, Oral Presentation, Acceptance Rate 4.4%, PDF)
  7. Z Qiao, X Qin, Y Zhou*, F Yang, W Wang. "Gaussian Constrained Attention Network for Scene Text Recognition." ICPR, 2020. (CCF-C, PDF)
  8. Y Zhang, C Liu, Y Zhou*, W Wang, W Wang, Q Ye. "Progressive Cluster Purification for Unsupervised Feature Learning." ICPR, 2020. (CCF-C, PDF)
  9. Y Zhou, Y Wang, J Cai, Y Zhou, Q Hu, W Wang. "Expert Training: Task Hardness Aware Meta-Learning for Few-Shot Classification." arXiv preprint, 2020. (PDF)
2019
  1. X Qin, Y Zhou*, D Yang, W Wang. "Curved Text Detection in Natural Scene Images with Semi- and Weakly-Supervised Learning." ICDAR, 2019. (CCF-C, PDF)
  2. Y Chen, Y Zhou*, D Yang, W Wang. "Constrained Relation Network for Character Detection in Scene Images." PRICAI, 2019. (CCF-C, PDF)
  3. N Jiang, Y Zhang, D Luo, C Liu, Y Zhou, Z Han. "Feature Hourglass Network for Skeleton Detection." SkelNetOn@CVPR, 2019. (PDF)
2016&Pre
  1. X Yun, Y Wang, Y Zhang, Y Zhou . “A Semantics Aware Approach to the Automated Network Protocol Identification.” TON, 2016. (SCI一区, CCF-A)
  2. X Wang, Z Sun, W Zhang, Y Zhou, Y. Jiang. “Matching User Photos to Online Products with Robust Deep Features.” ICMR, 2016. (CCF-B)
  3. Y Zhou, X Yang, Y Zhang, X Xu, Y Wang, X Chai, W Lin. “Unsupervised Adaptive Sign Language Recognition based on Hypothesis Comparison Guided Cross Validation and Linguistic Prior Filtering.” Neurocomputing, 2015. (SCI一区, CCF-C)
  4. W Lin, Y Zhang, J Lu, B Zhou, J Wang, Y Zhou. "Summarizing Surveillance Videos with Local-Patch-Learning-Based Abnormality Detection, Blob Sequence Optimization, and Type-Based Synopsis." Neurocomputing, 2015. (SCI一区, CCF-C)
  5. F Yin, X Chai, Y Zhou, X Chen. “Weakly Supervised Metric Learning towards Signer Adaptation for Sign Language Recognition.” BMVC, 2015. (CCF-C)
  6. F Yin, X Chai, Y Zhou, X Chen. “Semantics Constrained Dictionary Learning for Signer-Independent Sign Language Recognition.” ICIP, 2015. (CCF-C)
  7. H Wang, X Chai, Y Zhou, X Chen. “Fast Sign Language Recognition Benefited from Low Rank Approximation.” FG, 2015. (CCF-C)
  8. Y Zhou, S Liu, Y Zhang, Y Wang, W Lin. “Perspective Scene Text Recognition with Feature Compression and Ranking.” IWRR@ACCV, 2014. (Best Paper Award)
  9. Y Lin, X Chai, Y Zhou, X Chen. “Curve Matching from the View of Manifold for Sign Language Recognition.” FSLCV@ACCV, 2014.
  10. Y Zhou, Y Zhang, J Xiao, Y Wang, W Lin. “Visual Similarity Based Anti-Phishing with the Combination of Local and Global Features.” TrustCom, 2014. (CCF-C, Acceptance Rate 31.7%, Oral Presentation)
  11. S Liu, Y Zhou, Y Zhang, Y Wang, W Lin. “Text Detection in Natural Scene Images with Stroke Width Clustering and Superpixel.” PCM, 2014. (CCF-C)
  12. K Huang, Z Zhang, Y Chen, W Lin, Y Zhou, D Jiang, C Yao. “Improved Human Head and Shoulder Detection with Local Main Gradient and Tracklets-Based Feature.” APSIPA ASC, 2014.
  13. Y Zhou, W Lin, H Su, J Wu, J Wang, Y Zhou. “Representing and Recognizing Motion Trajectories: A Tube and Droplet Approach.” ACM MM, 2014. (CCF-A, Short Paper)
  14. Y Zhou, S Liu, Y Zhang, Y Wang, W Lin. “Text Localization in Natural Scene Images with Stroke Width Histogram and Superpixel.” APSIPA ASC, 2014.
  15. Y Sang, Y Zhang, Y Wang, Y Zhou, T Xu. “A Segmentation Pattern Based Approach to Automated Protocol Identification.” PDCAT, 2014. (CCF-C)
  16. Y Zhou, X Yang, W Lin, Y Xu, L Xu. “Hypothesis Comparison Guided Cross-Validation for Unsupervised Signer Adaptation.” ICME, 2011. (CCF-B) 
  17. N Xu, W Lin, Y Zhou, Y Chen, Z Chen, H Li, "A New Global-Based Video Enhancement Algorithm by Fusing Features of Multiple Region-of-Interests." VCIP, 2011.
  18. L Xu, S Kwong, T Zhao, Y Zhou. “Priority Pyramid Based Bit Allocation for Multiview Video Coding.” VCIP, 2011.
  19. Y Zhou, X Chen, D Zhao, H Yao, W Gao. “Adaptive Sign Language Recognition with Exemplar Extraction and MAP/IVFS.” IEEE SPL, 2010.(SCI二区, CCF-C)
  20. Y Zhou, X Chen, D Zhao, H Yao, W Gao. “Mahalanobis Distance Based Polynomial Segment Model for Chinese Sign Language Recognition.” ICME, 2008. (CCF-B, Oral Presentation, Acceptance Rate about 20%)
  21. Y Zhou, W Gao, X Chen, L Zhang, C Wang. “Signer Adaptation Based on Etyma for Large Vocabulary Chinese Sign Language Recognition.” PCM, 2007. (CCF-C)

      竞赛奖项

      1. ACM MM 2021, Best Paper Candidate, 5篇/1942篇, 2021
      2. CSIG 2022票据识别与分析挑战赛,冠军,2022
      3. CVPR DocVQA 2020,任务1第3名,2020
      4. 中国人工智能·多媒体信息识别技术竞赛,手写/印刷文本OCR两项高校组冠军,2019
      5. ICDAR ReCTS 2019,“字符识别”、“端到端识别”高校组第3名,2019
      6. CVPR SkelNetOn 2019,1项第2名,2019
      7. ACCV IWRR 2014 最佳论文奖,2014 

      专利软著

      发明专利

      1. 面向场景文本检测的文本检测器训练方法及文本检测方法, 2022, 受理号:202210492865.9
      2. 一种成本高效的场景文字检测方法及系统, 2021, 受理号: 202111295077.2
      3. 单阶段3D点云目标检测方法及装置、计算机设备、介质, 2021, 受理号:202111271651.0
      4. 基于并行迭代模仿解码的场景文字识别系统及方法, 2021, 受理号:202111026162.9
      5. 文本视觉问答方法和装置, 2021, 受理号:202111186856.9
      6. 基于多层感知机掩膜解码器的文字检测系统及方法, 2021, 受理号:202111034219.X
      7. 一种面向场景图像中任意形状邻近文本的检测系统及方法, 2021, 受理号:202111004566.8
      8. 一种基于密集语义对比的自监督视觉模型预训练方法, 2021, 受理号:202110988818.9
      9. 基于全卷积角点修正网络的多向场景文字检测方法及装置, 2021, 受理号:202110235490.3
      10. 一种多结构多任务深度神经网络及其训练、分类方法, 2020, 受理号:202011040925.0
      11. 基于语义强化编码器解码器框架的场景文字识别方法,2020,受理号:202010416704.2
      12. 一种基于自训练的文本检测器训练方法及系统,2020,受理号:202010428815.5
      13. 基于高斯约束注意力机制网络的场景文字识别方法及系统,2020,受理号:202010767079.6
      14. 基于完形填空任务的视频自监督学习方法,2019,受理号:201911348018.X
      15. 一种基于半监督与弱监督学习的曲形场景文字检测方法,2019,受理号:201910720688.3
      16. 基于受限注意力模型的字符检测网络训练方法、字符检测方法和字符检测器,2019,受理号:201910614874.9
      17. 一种基于特征压缩与特征选择的歪斜场景文字识别方法及系统, 2015, 受理号: 201510014950.4
      18. 一种基于三分类器协同训练学习的网络协议识别方法及系统, 2014, 受理号: 201410575510.1
      19. 基于直方图和超像素的场景图像文字检测方法及系统, 2014, 受理号: 201410168244.0
      20. 一种基于语义敏感的网络协议识别方法及系统, 2014, 受理号: 2014111800506700
      21. 一种自动检测疑似仿冒网站的方法及系统, 2013, 受理号: 201310395429.0
      22. 一种未知网络协议识别方法及系统, 2013, 受理号: 201310189079.2
      23. 一种用户隐私信息保护方法及系统, 2013, 受理号: 201310722437.1

      软件著作权

      1. 基于数据手套的中国手语识别软件系统,2009,软著登记号:2009SR02392

      科研项目

      • 印章识别及通用文字识别,主持,企业横向项目,2022.07-2024.06
      • 场景文字检测识别引擎,主持,企业横向项目,2022.01-2023.12
      • 噪声及低分辨率条件下的图像文本识别技术研究,主持,企业横向项目,2021.04-2022.10
      • 面向媒体融合与传播的富媒体信息智能提取技术,主持,国家重点实验室开放课题,2020.08-2021.07
      • 多媒体数据分析系统,主持,国家级,2018.09-2019.08
      • 实时数据检测分析系统, 主持, 国家级, 2017.10-2020.09
      • 基于云化平台的仿冒网站检测微引擎技术研究, 主持, 国家级, 2014.06-2017.06
      • 基于多示例学习和半监督学习的手势语识别研究, 主持, 国家级, 2014.01-2016.12
      • 多媒体内容取证方法研究, 参与, 国家级, 2013.01-2016.12
      • 海量信息分析系统, 参与, 国家级, 2014.06-2016.06
      • 文字图像中特定光学字符的快速检测方法研究, 主持, 市地级, 2013.06-2014.06
      • 面向复杂动态背景和可变环境的多模态手势语识别研究, 主持, 国家级, 2011.06-2012.06

      学生指导

      *含与王伟平研究员、中国传媒大学张远教授联合指导学生

      1. 杨东宝,助理研究员,2020级在职博士生,在读,发表期刊会议论文10余篇(含PR、TOMM等);
      2. 秦绪功,2017级博士生,一作ACM MM-21、ICASSP-21、ICDAR-19,入职南京理工大学(教职);
      3. 陈语地,2017级硕士生,一作ICPR-20、PRICAI-19,3项国内外竞赛前三名,入职抖音;
      4. 峙,2018级硕士生,一作ACM MM-21 Best Paper Candidate、CVPR-20、ICPR-20,入职好未来(SSP Offer);
      5. 罗德昭,2018级硕士生,一作AAAI-20 Oral、TOMM-22,QMUL龚少刚教授博士生;
      6. 曾港艳,2018级硕博生,一作ACM MM-21 Oral、PRICAI-21,在读;
      7. 李晓倪,2019级硕士生,一作ACM MM-21 Oral、PR-22、ICANN-21,入职北京银行;
      8. 过友辉,2019级硕士生,一作ICME-22、ICANN-21,入职科大讯飞(飞星计划
      9. 威,2020级硕士生,一作ACM MM-22,CSIG 2022票据识别与分析挑战赛冠军,2022在读;
      10. 波,2020级硕士生,一作ACM MM-22,在读;
      11. 谨,2020级硕士生,一作ACM MM-22,在读;