Wen GAO（President of the CCF）
| Visual Search by CDVS |
日時：3月10日（木）17:10-17:40
会場：特別会場（藤原洋記念ホール）
 | 【講演概要】Smart phone and surveillance systems have shown great potential for visual search. Emerging applications include landmark search, product search, CD (Compact Descriptor) or book cover search, location recognition, scene retrieval, car search, etc. There are at least two challenges for visual search, low latency transmission via wireless network connection, and high speed search in large image database at cloud server. A possible approach is to extract the visual feature at the capturing device, then sending that to cloud server for search. A practical issue is how to make visual search applications compatible across a broad range of devices and platforms. To solve the problem, we need a standard which can specify the feature set which is suitable for most applications.
In this talk, I will discuss CDVS: Compact Descriptor for Visual Search, CDVS, the standard created by ISO/IEC MPEG working group in 1995, known as ISO/IEC 15938-13, the part 13 of MPEG-7. CDVS uses feature descriptors instead of compressed images for transmitting and search, with high efficiency and acceptable search accuracy. To encode robust compact visual descriptors, advanced machine learning and data mining approaches have been employed to learn compact and discriminative properties from visual feature data, as well as to improve the search performance in dealing with a very large scale image database. In particular, a generative probabilistic model Fisher Kernel has been exploited in CDVS elegantly for scalable, compact and dense feature representation, which selectively aggregates local feature descriptors for efficient visual search.
I will also give some results on competitive and collaborative platform to evaluate the state-of-the-art visual search techniques and application solutions, where machine learning and data mining techniques have been shown to be the most promising approach to improve the performance and usability of visual search.
In the second part of my talk, I will briefly introduce the CCF, Chinese Computer Federation, the society for computer scientist and engineer, about its history, current situation, and future vision. 【略歴】Wen Gao received his Ph.D. degree in electronics engineering from the University of Tokyo in 1991. He is a professor at the Peking University since 2006. He serves as the vice president of NSFC from Feb. of 2013, and the president of CCF from Jan. 2016. Prof. Gao joined with the Harbin Institute of Technology from 1991 to 1995, as professor, department head of computer science. He was with Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS) from 1996 to 2005. During his career in CAS, he served as the Executive Director of ICT from 1998 to 1999, the executive president of Graduate School of CAS from 2000 to 2004, the vice president of University of Science and Technology China from 2000 to 2003. Prof. Gao works in the areas of multimedia and computer vision, including video coding, video analysis, multimedia retrieval, face recognition, multimodal interfaces, and virtual reality. He published six books and over 700 technical articles in refereed journals and proceedings in above areas. He earned many awards including six National Awards in Science and Technology Achievements. He has been featured by IEEE Spectrum in June 2005 as one of the "Ten To Watch" among China's leading technologists. He is a fellow of IEEE, a fellow of ACM, CCF Fellow, and a member of Chinese Academy of Engineering. |
