Welcome Professor Yuxin Peng to visit the HRI laboratory and give an academic report
Welcome Professor Yuxin Peng to visit the HRI laboratory and give an academic report
The atmosphere was enthusiastic, the students actively asked questions and listened very carefully!
Lecture time：September 5 (Thursday) 17:00
Lecture Venue：Peking University Shenzhen Graduate School C Building 126
Lecturer：Professor Yuxin Peng (Peking University)
Lecture Topic：Cross-Media Intelligence: Representation, Analysis and Application
Yuxin Peng , professor and doctoral supervisor of Peking University, leader in science and technology innovation of the National Leading Talent, leader of young and middle-aged science and technology innovation in the Ministry of Science and Technology, chief expert of the 863 project, chairman of the Expert Committee of China Artificial Intelligence Industry Innovation Alliance, and "Artificial Intelligence of the Chinese Academy of Engineering 2.0" Planning Expert Committee expert, director and deputy secretary-general of China Society of Image and Graphics. Main research directions: cross-media analysis and reasoning, image and video understanding and retrieval, and computer vision. In 2016, he won the first prize of Beijing Science and Technology Award (the first person to complete). Presided over more than 20 projects such as 863 and National Natural Science Foundation of China, and published more than 140 papers, including more than 60 papers of ACM/IEEE Trans and CCF. Participated 8 times in the international evaluation TRECVID video sample search competition organized by the National Institute of Standards and Technology NIST, and won the first place. The cross-media Internet content analysis and identification system developed by him has been applied to the Ministry of Public Security, the Ministry of Industry and Information Technology, and the State Administration of Radio and Television. Served as an editorial board member of "IEEE Transactions on Circuits and Systems for Video Technology" and other journals. Lecture summary: With the rapid development of multimedia and network technology, massive images, videos, texts and other cross-media data are growing rapidly. They are multi-source heterogeneous and interrelated, making data representation, information retrieval, knowledge discovery, and semantic reasoning face cross-cutting Challenges such as media and cross-data sources. How to learn from the cross-media characteristics of the human brain to recognize the external world across different sensory information such as vision, hearing, language, etc., is essential for improving the perception and cognitive ability and intelligence of computers. This report will introduce the tasks and goals of the cross-media analysis and reasoning technology in the "Artificial Intelligence 2.0" of the Chinese Academy of Engineering, and then focus on our relevant research progress, including fine-grained image classification, cross-media retrieval, text generation of images, and video description generation Wait.