Weiqiang Wang  Ph.D  Professor,  School of Computer Science and Technology, University of Chinese Academy of Sciences, CAS

Research Areas

My research interests involve machine learning, computer vison, and multimedia technology.


I received  the B.E. and M.E. degrees  in computer sciences  from Harbin Engineering University, in 1995 and 1998 respectively, and the Ph.D. degree in computer science from Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS), in 2001.


Work Experience

I am currently a professor with the college of computer and control engineering, Graduate University of Chinese Academy of Sciences (GUCAS),  a guest researcher with the institute of computing technology (ICT), CAS, and adjunct professor with Harbin Engineering University,  the members of IEEE, ACM.  I begain my career in the institute of computing technology  in May, 2001, as an assistant researcher, and was promoted as an associate researcher in Oct, 2001.  I jointed the GUCAS in Aug. 2003 as an associate professor, and was promoted as a professor in Jun., 2009. Meanwhile, I worked in the institute of Human-computer Interaction, Carnigie Mellon University (CMU), as a visiting scientist. By now I have led  the research work of some national projects, and published more than 100 papers.

Teaching Experience

    1. Computer vison I
    2. Image analysis and machine vision
    3. Machine vision and pattern recognition
    4. digital image processing




  • Lianlei Shan, Weiqiang Wang: MBNet: A Multi-Resolution Branch Network for Semantic Segmentation      Of Ultra-High Resolution Images. ICASSP 2022: 2589-2593


  • Ji Gan, Weiqiang Wang: HiGAN: Handwriting Imitation Conditioned on Arbitrary-Length Texts      and Disentangled Styles. AAAI 2021: 7484-7492
  •  Lianlei ShanXiaobin Li, Weiqiang      Wang: Decouple the High-Frequency and      Low-Frequency Information of Images for Semantic Segmentation. ICASSP 2021: 1805-1809


  • Ji Gan, Weiqiang Wang, Ke LuIn-air handwritten Chinese text recognition with temporal      convolutional recurrent network. Pattern Recognit. 97 (2020)


  • Ji Gan, Weiqiang Wang, Ke LuA new perspective: Recognizing online handwritten Chinese characters      via 1-dimensional CNN. Inf. Sci. 478: 375-390 (2019)
  • Yang Bai, Weiqiang Wang: ACPNet: Anchor-Center Based Person Network for Human Pose Estimation      and Instance Segmentation. ICME 2019: 1072-1077


  • Xiwen Qu, Weiqiang Wang, Ke LuJianshe ZhouIn-air handwritten Chinese character recognition with      locality-sensitive sparse representation toward optimized prototype      classifier. Pattern Recognit. 78: 267-276 (2018)
  • Ji Gan, Weiqiang Wang, Ke LuA Unified CNN-RNN Approach for in-Air Handwritten English Word      Recognition. ICME 2018: 1-6
  • Peirui Cheng, Weiqiang Wang:  A Multi-Oriented      Scene Text Detector with Position-Sensitive Segmentation. ICMR 2018: 152-159


  • Kang Shi, Weiqiang Wang: Sweeper: Design of the Augmented Path in Residual Networks. PCM (1) 2017: 811-821


  • Zhaochun Xu, Weiqiang Wang, Ke LuMQDF with a novel covariance matrix estimation and discriminant      LSRC, which is better for in-air handwritten Chinese character      recognition. ICIP 2016: 3274-3278
  • Xiwen Qu, Weiqiang Wang, Ke LuZhangjian JiHigh-order directional features and sparse representation based      classification for in-air handwritten Chinese character recognition. ICME 2016: 1-6
  • Shao Huang, Weiqiang Wang, Ke LuEgocentric hand detection via region growth. ICPR 2016: 639-644
  • Xiwen Qu, Weiqiang Wang, Ke LuIn-air handwritten Chinese character recognition using      discriminative projection based on locality-sensitive sparse      representation. ICPR 2016: 1137-1140


  • Ping Hu, Weiqiang Wang, Ke LuVideo text detection with text edges and convolutional neural      network. ACPR 2015: 675-679
  • Ning Xu, Weiqiang Wang, Xiwen QuRecognition of In-air Handwritten Chinese Character Based on Leap      Motion Controller. ICIG (3) 2015: 160-168
  • Ning Xu, Weiqiang Wang, Xiwen QuA discriminative classifier for in-air handwritten Chinese      characters recognition. ICIMCS 2015: 16:1-16:4
  • Zhangjian Ji, Weiqiang Wang, Ke LuRobustly tracking objects via multi-task kernel dynamic sparse      model. ICIP 2015: 266-270
  • Shao Huang, Weiqiang Wang: Retrieving images combining saliency detection with IRM. ICIP 2015: 517-521
  • Xiwen Qu, Weiqiang Wang, Ke LuNing XuIn-air handwritten Chinese character recognition using multi-stage      classifier based on adaptive discriminative locality alignment. ICIP 2015: 4793-4797
  • Xiwen QuNing Xu, Weiqiang Wang, Ke LuSimilar handwritten Chinese character recognition based on adaptive      discriminative locality alignment. MVA 2015: 130-133
  • Ning Xu, Weiqiang Wang, Xiwen QuOn-line Sample Generation for In-air Written Chinese Character      Recognition Based on Leap Motion Controller. PCM (1) 2015: 171-180


  • Zhangjian Ji, Weiqiang Wang: Robust object tracking via multi-task dynamic sparse model. ICIP 2014: 393-397
  • Zhike Zhang, Weiqiang Wang, Ke LuVideo Text Extraction Using the Fusion of Color Gradient and      Log-Gabor Filter. ICPR 2014: 2938-2943
  • Shao Huang, Weiqiang Wang: A Filtration Strategy Based on FD-CRF for Image Matching. ICPR 2014: 3286-3291


  • Chi Zhang, Weiqiang Wang: Object-level saliency detection based on spatial compactness      assumption. ICIP 2013: 2475-2479
  • Zhangjian Ji, Weiqiang Wang, Ke LuExtract foreground objects based on sparse model of spatiotemporal      spectrum. ICIP 2013: 3441-3445
  • Zhangjian Ji, Weiqiang Wang, Ke LvForeground Detection Utilizing Structured Sparse Model via l1, 2      Mixed Norms. SMC 2013: 2286-2291
  • Zhike Zhang, Weiqiang Wang: A Novel Approach for Binarization of Overlay Text. SMC 2013: 4259-4264


  • Chi Zhang, Weiqiang Wang: A robust and efficient shot boundary detection approach based on      fisher criterion. ACM Multimedia 2012: 701-704



  • Jian LinYi Sun, Weiqiang      Wang: Violence Detection in Movies      with Auditory and Visual CuesCIS 2010: 561-565
  • Shugao Ma, Weiqiang Wang: Effective camera motion analysis approach.       ICNSC 2010: 111-116


  • Jian Lin, Weiqiang Wang: Weakly-Supervised Violence Detection in Movies with Audio and Video      Based Co-training. PCM 2009: 930-935



  • Weiqiang Wang, Libo FuWen GaoText Segmentation in Complex Background Based on Color and Scale      Information of Character Strokes. PCM 2007: 397-400



  • Libo Fu, Weiqiang Wang, Yaowen ZhanA Robust Text Segmentation Approach in Complex Background Based on      Multiple Constraints. PCM (1) 2005: 594-605