In this talk, we first briefly discuss an efficient and scalable retrieval method for Internet video content that we developed at Carnegie Mellon University. We implemented E-Lamp Lite, the first of its kind content-based search engine for Internet videos. After that, we will discuss our recent research on vision+language and weakly-supervised deep learning at Google AI. A few interesting problems such as video question answering and video understanding will be discussed to showcase our deep learning research.
Lu Jiang is a research scientist at Google Cloud AI. He received a Ph.D. in Artificial Intelligence (Language Technology) from the Carnegie Mellon University in 2017. Lu’s primary interests lie in the interdisciplinary field of Multimedia, Machine Learning, and Computer Vision, which, specifically, include video understanding, language+vision, and weakly supervised learning. He is the recipient of Yahoo Fellow and Erasmus Mundus Scholar. He received the best poster award at IEEE Spoken Language Technology and the best paper nomination twice at ACM International Conference on Multimedia Retrieval. He was the key contributor to IARPA project Aladdin in CMU. The system achieves the top performance in TRECVID 2013-2015 organized by the National Institute of Standards and Technology (NIST). He serves on the program committee of premier conferences such as ACM Multimedia, AAAI, IJCAI, and CVPR and the journal reviewer of JMLR, TPAMI, TMM, CVIU etc.
蒋路，谷歌科学家，谷歌云人工智能(李飞飞)团队创始成员。曾于2017年获得卡内基梅隆大学人工智能（语言技术）博士学位。曾于2008年2011年获得西安交通大学工学学士与硕士学位。长期致力于计算机视觉，机器学习和多媒体的交叉领域研究。他是Yahoo Fellowship和欧盟的Erasmus Mundus Scholarship的获得者。作为核心成员获得美国国家标准总局（NIST）举办的多项大赛中获得冠军。近年来在包括 NIPS, ICML, CVPR, ECCV, MM, AAAI, IJCAI 等发表论文20余篇。他是ACM Multimedia, AAAI, IJCAI, CVPR 等会议的技术程序委员，JMLR, TPAMI, TMM, CVIU等期刊的审稿人。工作曾经获得IEEE SLT的最佳论文（best poster）和ACM ICMR的最佳论文提名（best paper candidate）。