谷歌(Google)软件工程师,主要致力于视频编码与视频通信的算法设计与实现。曾在美国贝尔实验室(Bell Labs), 诺基亚研究中心(Nokia Research Center), 以及惠普实验室(HP Labs)等处从事理论算法研究,后转入工业届,尤其参与了如下视频通话产品的设计与推出:苹果(Apple)的FaceTime,谈客 (TangoMe)Video Calls,以及 谷歌眼镜(Google Glass)专属Hangouts Video Calls。目前在谷歌从事AV1视频编码的标准制作与产品优化,以及图像、视频相关领域新技术的探索。
In this talk, we will present the argument in favor of an open source, royalty-free video codec that will keep pace with the evolution of video traffic. Additionally, we argue that the availability of a state-of-the-art, royalty-free codec levels the playing field, allowing small content owners and application developers to compete with the larger companies that operate in this space. This will ultimately result in a richer and more diverse internet. We will then introduce the Alliance for Open Media (AOM), a consortium with a lineup of tech giants including Google, Adobe, Amazon, AMD, Broadcom, Cisco, Google, Hulu, IBM, Intel, Microsoft, Mozilla, nVIDIA, Netflix, etc. 25+ board members, many of which own patent leading Internet companies. AOM is dedicated to developing open source, royalty free, new generational media formats, codecs and technologies. We will finally elaborate the showcase of YouTube which has increasingly heavily on VP9 - the direct ancestor of the AOM video codec.
Google started the WebM Project in 2010 to develop open source, royalty-­free video codecs designed specifically for media on the Web. The second generation codec released by the WebM project, VP9, ­is currently served by YouTube, and enjoys billions of views per day. Realizing the need for even greater compression efficiency to cope with the growing demand for video on the web, the WebM team embarked on an ambitious project to develop a next edition codec AV1, in a consortium of major tech companies called the Alliance for Open Media (AOM), that achieves at least a generational improvement in coding efficiency over VP9. In this talk, we focus primarily on the new tools developed by AV1, including the tools and coding modes for the prediction of pixel blocks that improve intra, inter and combined inter-intra prediction, as well as new paradigms for transform, interpolation filtering, coefficient entropy-coding, super-resolution, and restoration filtering.