报告嘉宾：严骏驰（IBM中国研究院，上海）
报告时间：2016年3月30日晚21:00（北京时间）
报告题目：Consistency-driven Multiple Graph Matching
报告题目：一致性驱动的多图匹配模型和算法（中文） 主持人： 程明明（南开） 报告相关文献列表：
Junchi Yan, M. Cho, H. Zha, X. Yang, S. Chu, Multi-Graph Matching via Affinity Optimization with Graduated Consistency Regularization, IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), DOI10.1109/TPAMI.2015.2477832
Junchi Yan, J. Wang, H. Zha, X. Yang, S. Chu, Consistency-Driven Alternating Optimization for Multigraph Matching: A Unified Approach, IEEE Transactions on Image Processing (T-IP), 24 (3), 994-1009, 2015
Junchi Yan, Y. Li, W. Liu, H. Zha, X. Yang, S. Chu, Graduated consistency-regularized optimization for multi-graph matching, Proceedings of the13th European Conference on Computer Vision (ECCV’14), Zurich, Switzerland, September 6-12, 2014, p.407-422
Junchi Yan, Y. Tian, H. Zha, X. Yang, Y. Zhang, S. Chu, Joint optimization for consistent multiple graph matching, Proceedings of IEEE International Conference on Computer Vision (ICCV’13), Sydney, Australia, December 1-8, 2013, p.1649-1656 报告摘要：
Graph matching solves the problem of finding node correspondence between two or multiple graphs structures. It is one of the fundamental problems in computer science, and also plays an important role in computer vision and pattern recognition. This talk will start with an introduction for two-graph matching, specifically with a predefined affinity matrix, and further presents two recent methods for joint multiple graph matching. One approach involves a general alternating optimization procedure which handles with both factorized and non-factorized forms of the affinity matrix; the other relates to a matching affinity boosting strategy and meanwhile is gradually regularized by cross-graph matching consistency indicators. 报告人简介：
严骏驰于2011年3月硕士毕业于上海交通大学模式识别与智能系统专业，随后加入IBM中国研究院，先后担任Researcher、Staff Researcher、Research Staff Member。在此期间，于2012年起在职攻读博士学位，并于2015年从上海交通大学信息与通信工程获得博士学位。自2015年10月起，作为华东师范大学在站博士后，主持IBM与华东师范大学关于企业市场推荐系统的联合研究项目。他的主要研究兴趣为计算机视觉、模式识别与机器学习相关应用，近年来在CVPR/ICCV/ECCV/AAAI/IJCAI/TPAMI/TIP上发表第一作者论文9篇，并担任CVPR/ICCV/ECCV/IJCAI/TPAMI/TIP/PR审稿人。他是IBM Master Inventor和ACM中国优秀博士论文提名奖获得者。
Upon receiving his M.S. degree from Shanghai Jiao Tong University in the major of Pattern Recognition and Intelligent System in March 2011, Junchi Yan joined IBM Research – China and successively held research position as a Researcher, Staff Researcher, and Research Staff Member. In 2012, he started to pursue his part-time PhD and received PhD from Shanghai Jiao Tong University in the major of Information and Communication Engineering in 2015. He is now leading the Shared Union Research project between IBM and East China Normal University on the project of Business-to-Business recommender system and works as a postdoc to drive this initiative. His main research covers computer vision, pattern recognition, and machine learning applications. He has first-authored 9 papers in CVPR/ICCV/ECCV/AAAI/IJCAI/TPAMI/TIP. He serves as reviewer or technical program committee member for CVPR/ICCV/ECCV/IJCAI and TPAMI/TIP, Pattern Recognition. He is an IBM Master Inventor and is the recipient of the Nomination Award of 2015 ACM China Doctoral Dissertation Award.
报告材料[Slides] | 