Reims Champagne-Ardenne University | Centre de Recherche en STIC +33.(0)

Xiaofang Wang

Department : Informatique


IUT de Reims-Châlons-Charleville rue des crayères BP 1035 51687



  • PhD, Ecole centrale de Lyon (ECL), France — computer vision, Sep. 2011 -  March 2015, Title: Unsupervised image segmentation and multiple object tracking based on graph
  • MS, Central south university, ChangSha, China— Biomedical Engineering, Sep. 2009 -  Jun 2011, Title: Liver CT image segmentation based on active contour models
  • BS, Centrale south university, Changsha, China Biomedical Engineering, Sep. 2005 -  Jun.  2009. Topic:  Medical image segmentation


  • 2019, Research Engineer at Siradel
  • 2018, Post-doc at Inria Rennes
  • 2017, Post-doc at Ecole Centrale de Lyon
  • 2016, Associated Professor at Ecole Centrale de Lyon
  • 2015, Associated Professor at  Ecole Centrale de Lyon


  • 2018, superviser one stage of third-year engineers 
  • 2015-2017, co-supervise a PhD these in Shanghai Jiaotong University, China 
  • 2014, Invite and helped install Prof. Zhao Yuqian to visit Liris in Ecole Centrale de Lyon
  • INF-TC1: Introduction to algorithms (1st year Ecole Centrale de Lyon, TD/TP, 84h), Fall 2015 and Spring 2016.
  • INF-TC2: Object-oriented programming (1st year Ecole Centrale de Lyon, TD/TP, 63h), Fall 2015 and Spring 2016.
  • INF-TC3: Web and Database Project (1st year Ecole Centrale de Lyon, TD/TP, 32h), Fall 2015 and Spring 2016.



  • 2011 -  2015,  ECL, Visen — tagging the visual content with semantic labels
  • 2017 -  2018, ECL, Pikaflex  picking and kitting objects with an automatic robot grasp system
  • 2018- 2019, INRIA, SafeCity— video understanding and surveillance system applying in key sites within different cities of France
  • 2020,  Uiniversity of Reims, Schedar -- Safeguarding the Cultural HEritage of Dance through Augmented Reality



Point in, Box out: Beyond Counting Persons in Crowds. 
Yuting liu, Miaojing Shi, Qijun Zhao, Xiaofang Wang. 
IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'2019), Link 
Developmental Bayesian Optimization of Black-Box with Visual Similarity-Based Transfer Learning. 
Maxime Petit, Amaury Depierre, Xiaofang Wang, Emmanuel Dellandrea, and Liming Chen. 
IEEE International Conference on Development and Learning (ICDL) and the International Conference on Epigenetic Robotics (EpiRob), 2018, Tokyo.  PDF 
Fusing Generic Objectness and Deformable Part-based Models for Weakly Supervised Object Detection [pdf]
Yuxing Tang, Xiaofang Wang, Emmanuel Dellandréa, Simon Masnou, Liming Chen
IEEE International Conference on Image Processing (ICIP), Paris, 2014. (Top 10%)
A graph-cut approach to image segmentation using an affinity graph based on ℓ0-sparse representation of features [pdf]
Xiaofang Wang,  Huibin Li and Charles-edmond Bichot and, Simon Masnou, Liming Chen
IEEE International Conference on Image Processing (ICIP), 2013. (Top 10%)
Graph-based image segmentation using weighted color patch [pdf]
Xiaofang Wang and Chao Zhu and Charles-edmond Bichot and, Simon Masnou
IEEE International Conference on Image Processing (ICIP), 2013. 
Sparse Coding and Mid-Level Superpixel-Feature for ℓ0-Graph Based Unsupervised Image Segmentation[pdf]
Xiaofang Wang, Huibin Li, Simon Masnou, Liming Chen 
Computer Analysis of Images and Patterns. Springer Berlin Heidelberg (CAIP), 2013. 
An improved non-local cost aggregation method for stereo matching based on color and boundary cue [PDF]
Dongming Chen, Mohsen Ardabilian, Xiaofang Wang, Liming Chen
IEEE International Conference on Multimedia and Expo (ICME), 2013.  
Research advances and prospects of mathematical morphology in image processing
Zijuan Yu, Yuqian Zhao, Xiaofang Wang 
IEEE Conference on Cybernetics and Intelligent Systems (pp. 1242-1247).  2008


Discriminative and geometry aware unsupervised domain adaptation. 
Lingkun Luo, Liming Chen, Shiqiang Hu, Ying Lu, Xiaofang Wang 
Submitted to IEEE Transactions on Cybernetics (TCYB), Link  
Accpeted 2019
Visual and Semantic Knowledge Transfer for Large Scale Semi-Supervised Object Detection 
Yuxing Tang, Josiah Wang, Xiaofang Wang, Boyang Gao, Emmanuel Dellandrea,and Liming Chen. 
IEEE transactions on pattern analysis and machine intelligence (T-PAMI), 2018, vol. 40, no 12, p. 3045-3058. Link 
A Global/Local Affinity Graph for Image Segmentation [link]
Xiaofang Wang, Yuxing Tang, Simon Masnou, Liming Chen
IEEE Transactions on Image Processing (TIP), vol. 24(4), pp.1399-1411, 2015
Weakly Supervised Learning of Deformable Part-Based Models for Object Detection via Region Proposals [pdf]
Yuxing Tang, Xiaofang Wang, Emmanuel Dellandréa, Liming Chen
IEEE Transactions on Multimedia, Institute of Electrical and Electronics Engineers (TMM),vol.19(2), pp. 393-407,2016
Active Colloids Segmentation and Tracking  [PDF]
Xiaofang Wang, Boyang Gao, Simon Masnou, Liming Chen, 
Pattern Recognition (PR) vol.60, pp. 177-188, 2016 
Retinal vessels segmentation based on level set and region growing [PDF]
Yu Qian Zhao and Xiao Hong Wang and Xiaofang Wang, and Frank Y Shih. 
Pattern Recognition (PR) vol.47(7), pp. 2437-2446,2014 
A Unified Framework for Interactive Image Segmentation via Fisher Rules. 
Lingkun Luo, Xiaofang Wang, Shiqiang Hu, Xing Hu, Huanlong Zhang,Yaohua Liu 
The Visual Computing, 2018 Link 
Interactive image segmentation based on samples reconstruction and FLDA. 
Lingkun Luo, Xiaofang Wang, Shiqiang Hu, Xin Hu, Liming Chen 
Journal of Visual Communication and Image Representation 43 (2017): 138-151. Link 
Level-set Method Based On Global and Local Regions For Image Segmentation [PDF]
Yuqian Zhao, Xiaofang Wang, Frank Y.Shih, Gang Yu 
International Journal of Pattern Recognition and Artificial Intelligence, vol. 26(01), 2013.

My research activities are centered on machine learning. I have studied different theories in machine learning and various topics in computer vision both in 2D and 3D.  The research contents can be images (medical, natural, aerial imagery, multi-band remote imagery, detph images), videos etc.  they are: 

  • Human Performance Capture and 3D pose and shape estiamtion
  • Unsupervised learning:  graphical models and spectral clustering for image segmentation and multiple objects tracking
  • Weakly supervised learning: transfer image-level information to object-level detection and localization based on deep learning
  • Supervised learning :  fully supervised learning based on deep learning on semantic segmentation 
  • Domain adaptation:  transfer image-level knowledge  learned in a given domain to another domain for image segmentation
Cette liste bibliographique est récupérée automatiquement depuis HAL
Cette liste bibliographique est récupérée automatiquement depuis HAL