美国Texas State University的Byron Gao教授和Pennsylvania State University的Yanxi Liu教授将分别来我系访问并作学术报告。报告时间，地点，及内容如下。如有老师或同学要和他们单独会见，请与张仲非老师联系。
Byron的研究方向时网络检索和数据挖掘。Yanxi的研究方向时机器学习及其在计算机视觉，图形学，和医学图象上的应用。
时间：五月二十七日下午2时
地点：信电系225（会议室）
报告人：Byron Gao
On Several Untypical Information Retrieval and Web Search Paradigms
In the BoBo project, we study the two-box search paradigm that features two input boxes on the search interface. Besides a search box taking search terms as in normal search engines, a domain box is used to take domain knowledge in the form of keywords. As search terms are inherently ambiguous, domain terms can be optionally used to route search results towards a user-intended domain.
In the Cager project, we study cross-page web search. Existing search engines have page as the unit of retrieval of information. Generally, given a query as a set Q of keywords, they return a ranked list of web pages, each containing Q. However, quite often, users wish to have what we call "cage" as the unit of retrieval. A cage, crossing multiple pages, is a set of closely related web pages that collectively contain Q.
In the Rant project, we try to provide a framework for mass-collaboration-based social search. Google (login) and Microsoft (U Rank) are experimenting on search engines that allow people to organize, edit and annotate search results, as well as share information with others. Currently there is no much systematic research on how to interpret, store, preserve, and utilize user preferences.
－－－
时间：六月一日上午10时
地点：教七102室
报告人：Yanxi Liu
Machine Learning for Computational Regularity
-- Symmetry Discovery from Real World Patterns
We explore a formal and computational characterization of real world regularity using discrete symmetry groups (hierarchy) as a theoretical basis, embedded in a well-defined Bayesian framework. Our existing work on ‘Near-regular texture analysis and manipulation’ (SIGGRAPH 2004) and “A Lattice-based MRF Model for Dynamic Near-regular Texture Tracking” (TPAMI 2007) already demonstrate the power of such a formalization on a diverse set of real problems, such as texture analysis, synthesis, tracking, perception and manipulation in terms of regularity. Symmetry and symmetry group detection from real world data turns out to be a very challenging problem that has been puzzling computer vision researchers for the past 40 years (CVPR 2008). Our novel formalization will lead the way to a more robust and comprehensive algorithmic treatment of the whole regularity spectrum, from regular (perfect symmetry), near-regular (approximate symmetry), to various types of irregularities. The proposed method will be justified by several real world applications in computer vision, computer graphics and biomedical image analysis applications such as deformed lattice detection and tracking (PAMI 2009), gait recognition (ECCV2002,CVPR2007), grid-cell clustering (Neurocomputing 2007), symmetry of dance (SIGGRAPH ASIA 2009), automatic geo-tagging (CVPR2008), shape matching and retrieval (ICCV 2009), and image de-fencing (CVPR2008,ICCV2009).