讲座1:Multimedia Analytics for Insight in Large Image Collections

演讲人：Marcel Worring

讲座时间：2014-05-15， 星期四，下午 4点-5点

讲座地点: 信息楼417

讲座内容:

Image collections are a tremendous source of information. Yet due to the semantic gap it is difficult to get access to their content and properly employ their context such as tags, geolocation, and other metadata. To move forward, we propose multimedia analytics solutions. These methods bring human experts and machine algorithms together in a synergetic manner where the machine does the bulk processing and the expert deals with the difficult decisions. Advanced interactive visualizations provide the means to let the expert understand the results of automatic processing and interactively provide the system with further guidance. Our solutions build upon our state-of-the-art visual concept detectors. From there we consider methods for interactively summarizing and exploring the collection. In particular we propose multimedia pivot tables, an extension of the pivot table reports found in spreadsheets and show how they provide highly flexible and versatile explorative capabilities. They provide a step towards methods to get valuable insight in large image collections.

演讲人简历：

Marcel Worring is associate professor at the Intelligent Systems Lab Amsterdam. He is co-initiator and associate director of the recently established Data Science Research Center Amsterdam bringing together the leading research institutes in Amsterdam (CWI, VU, UvA, HvA) on the whole data science chain from data acquisition, storage and distributed processing, to analysis, retrieval, and visualization. His research interests are in multimedia analytics, leveraging synergy in human-computer processes. He has published over 170 papers in refereed journals and conferences. He is general co-chair of ACM Multimedia 2016 in Amsterdam, was program co-chair for ACM Multimedia 2013 and ICMR 2013, and co-initiator and co-organizer of the VideOlympics 2007-2009. He was associate editor of Pattern Analysis and Applications and IEEE Transactions on Multimedia and currently is associate editor of ACM TOMCCAP.  He was guest editor for various special issues. He is senior member of both ACM and IEEE. He was a visiting researcher at Yale University (1992) and University of California San Diego (1998).

讲座2: Multimedia Question-Answering by Visual Search

演讲人：Chong-Wah Ngo

讲座时间：2014-05-15， 星期四，下午5点-6点

讲座地点: 信息楼417

讲座内容:

In real-life, it is easier to provide a visual cue when asking a question about a possibly unfamiliar topic, for example, asking the question, “Where was this crop circle found?”. Providing an image of the instance is far more convenient than texting a verbose de1111ion of the visual properties, especially when the name of the query instance is not known. Nevertheless, having to identify the visual instance before processing the question and eventually returning the answer makes multimedia-based question-answering technically challenging. In this talk, I will introduce a system recently built based on visual instance search to address this problem. The system performs search in a million-scale image dataset in order to mine the name of instance, and then use the name to search for similar question-answer pairs from Yahoo! Answer. The system is evaluated on a wide range of visual instances (e.g., fashion, art, food, pet, logo, and landmark) over various QA categories (e.g., factoid, definition, how-to, and opinion).

演讲人简介：

Chong-Wah Ngo is an associate professor in the Dept. of Computer Science at the City University of Hong Kong. He received his PhD in Computer Science from Hong Kong University of Science & Technology (HKUST), and MSc and BSc, both in Computer Engineering, from Nanyang Technological University of Singapore. Before joining City University of Hong Kong, he was a postdoctoral scholar in Beckman Institute at the University of Illinois in Urbana-Champaign (UIUC). He was also a visiting researcher of Microsoft Research Asia. His recent research interests include large-scale multimedia information retrieval and mining, video summarization and browsing. He is the founding leader of video retrieval group (VIREO) in City University. He is currently the associate editor of IEEE Transactions on Multimedia. He serves as the technical program committees of various multimedia conferences, including as the conference co-chair of ICMR 2015, PCM 2014, the program co-chair of ICMR 2012, MMM 2012, and the area co-chair of ACM Multimedia 2012 and ICME 2013. He served as the chairman of ACM (Hong Kong Chapter) during 2008-2009.