From the evolution of codec to WebRTC, real-time Internet technology standards, affecting the strategic and technological selection of each company in the industry. Although H.264 is still the most widely mainstream standard, HEVC, VP9, AV1 is more advantageous at coding efficiency.
In the past year, H.265 / HEVC officially launched a year of the 5th year, although Apple added it to HLS, but which opportunities will it face? AV1 released in the form of a draft this year, is a concern in the industry, what is the actual application effect? What practical cases? Domestic AVS standards, what are the advantages of other codec technology? Which challenges will be faced after WebRTC 1.0?
About codec and WebRTC's latest technical practices and evolution trends, at this RTC 2018 real-time Internet conference, you will hear the most official answer.
01
Four highlights, no virtual
Highlight 1: Acoustic signal acquisition, processing and reconstruction in urban communication and intelligent interaction
Voice communications and human-speaking voice interactions were originally two different areas, but with the development of technology, these two areas have high intersections in their own technical needs or in the face of applications. Especially In terms of acoustic signal acquisition, processing and reconstruction, both require high-fidelity, high-quality far field pickup, and reserve and reconstruct the signal and its spatial information under the complex scene.
Signal processing scientists will share the sound signal perception, handling and reconstruction of general processes and the main scientific problems to solve in this process, and the status quo of key technologies, and discussing in complex, far field pickup environments The main challenges faced by acoustic signal perception, acquisition, processing, transmission and reconstruction.
Highlight 2: New coding era, AVS2 audio and video standard evolution and application examples
AVS2 has begun to be applied to movies, television and video industries. And what is the key technology of AVS2, what is the relative advantage? What experience in application practice? What is the future plan? Many people may not know.
The length of the AVS Standards Committee audio group, test group and video group will not only answer these questions, but also share the latest developments in international / domestic new generation video coding standards, point cloud, light field and other emerging media coding, and Depth study in the field of video coding, etc.
Highlights 3: New generation video coding, dilemma and opportunities in interactive live services
Interactive live relative to on-demand (VOD) and traditional linear television (ground TV, cable television, IPTV) in play platform, background architecture, technical requirements, etc., there are many latitude differences. Based on these particularity, the upstream video coding industry has recently declined, interactive live platforms face unprecedented technical challenges for online coding formats. In fact, there are very few live broadcast platforms that deploy H.264 worldwide. At the same time, HEVC, VP9, AV1 have a very obvious coding efficiency advantage.
Twitch is currently 15 million, and the peak interactive live broadcast platform is more than 2.5 million. Chief R & D engineer from Twitch will analyze the compatibility of the playback platform from the front rear stage and the feasibility of high-graphic real-time coding, thus roughly outlined the forecast of the proposal in the next five years. In addition, it will also focus on the design of Switch_Frame in AV1, and further reducing live delay for Switch_Frame.
Highlight 4: WebRTC 1.0 and future evolution
In the past year, WebRTC implements unity on the browser, and has launched the industry standard WebRTC 1.0. The new version of the WebRTC version has begun. We also revealed some information before this. At this conference, members from Google's WebRTC Product Manager and WEBRTC Standards Committee will bring further sharing.
02
Who will share?
The technical experience of scientists in this signal processing area is rich, and we only share a part of the space.
He has successively engaged in research work in the field of signal processing, speech synthesis, speech recognition, and other fields in the Japan International Electrical Communications Basic Technology Institute (ATR) and Australia's Griffith University.
It is also engaged in the development of adaptive signal processing, arrays and MIMO signal processing and speech signal processing and communication in the United States.
He also served as a chief scientist of Wevoice. In 2010, I returned to China, I was selected for the third batch of national "Thousand Talents Program", and after joining Northwestern Industrial University, the Director and Chief Scientist of Intelligent Acoustics and Uribiting Communication Center, and some of the technologies developed have been successfully used for wireless communications, conference calls, Among the voice communication systems such as remote collaboration, smart speakers, vehicles.
I have received the best papers of the International IEEE Signal Processing Society, two won the Bell Lab Modular Team Award, two won the NASA Technical Innovation Award, has been published 12, published in the field of technical publications and meetings in the field of signal processing. Nearly 200 articles.
When the R & D team led by the Twitchch, the R & D team led by Twitch was responsible for the research and development of Twitch core video technology, the responsibility covers the live video transfer, ABR playback algorithm, multi-platform playback compatibility, picture quality, delay, etc.
Dr. Shen is also the inventor of the AllianceOfopenMedia video encoding protocol AV1, he published, and applied for more than 15 technical patents.
Prior to joining Twitch, Shen Qian's respectively, served in multiple digital TV equipment companies (GDMEDIWARE, AMBARELLA, HARMONIC, Ericssontv) and the start-up enterprise in the open cloud game industry. In these companies, he is dominated, participating in the development of multiple widely used H.264 encoding, transcoding, nonlinear editing, and real-time advertising insert products, and cloud game core technologies transmitted on the public Internet.
Professor Peking University School of Information Science and Technology, 2005, 2005, graduated from the Institute of Computational Technology, Chinese Academy of Sciences. From 2005 to 2007, he studied postdoctoral at the University of South California, and then went to Northern University. The main research direction is video coding and processing, more than 200 papers has been published, and more than 40 invention patents have been authorized. Served as IEeetransAncircuitsandsystemForvideTechnolgoy (TCSVT), JournalsuvisualCommunications (TCSVISUALCOMMUNICAON (TCSVISUALCOMMUNICATIONANDREPREPRETATION (JVCIR) Journal Society (AE), China Image Graphics Society, AVS video group combined group leader, etc. Since 2002, the development of a series of national standards of AVS1, AVS + and AVS2 has been involved in the formulation of national standards, and has been awarded the second prize of the National Technology Invention Award, the Second Prize of National Science and Technology Progress Award.
Dr. Pan Xingde, Dr. Beijing University of Posts and Telecom, Panorama Science & Scorpio K Ge Song, AVS Audio Group, Test Group, Joint Group. Study and application of long-term speech decoding technology, sound field technology and sound effect technology. Hosting or participating in the formulation of EVD, AVS and IEEP1857 and other standards, applying for nearly 100 invention patents in the field of audio technology, and is widely used by audio technology standards.
At present, the Chinese Panorama Sound Technology (WANOS) of Panorama Sound Technology has been widely used as a global two sets of panoramic sound technology standards, and has been widely used in film production and issuance and has gradually entered the field of network applications such as OTT TV. In addition to the Joint Group leader in the AVS audio group, the test group is jointly divided, and now serves as the IEEEVR audio standard convener, IEEE, AES, e-learning and acoustic members.
Chen Cheng's undergraduate graduated from Tsinghua University Automation, after the University of Iowa University, I received a doctoral degree in Iowa, and I was in Google. I was affiliated to the video compression core algorithm group. The R & D and software development of VP9 and AV1 video compression standards include AV1. The extension of the block filter is expanded, and the VP9 / AV1 encoding optimization, etc. based on relative distances. In addition to video compression technology, research interests also include image compression, machine learning algorithms and their applications in image and video.
Zoe Liu (Liu Yuxin) is the joint founder of Visionular (Micro-frame Technology), Chairman and Chief Scientist.
5 years before this, Zoe has served as a Software engineer of Google Chrome Media team and as an open source video codec standard AOM / AV1 core member participated in R & D and standard development.
She received bachelor, master's degree and Ph.D. degree at Tsinghua University, and received the second doctoral degree in Puerto University.
Regardless of the main contributor or technical person in charge, Zoe has a prominent contribution in the design and development of multiple audio and video products, including Apple Facetime, Tango Video Phone, Google Glass Video Phone, etc. Zoe has many years of innovative research experience in multiple famous research labs, including Bell Lab, Nokia Research Center, Solar Microprocessor Center Lab, HP Lab.
Daniel C. Burnett has worked for ten years in the computer standard, as the editor of Peerconnection and GetUserMedia W3C WebRTC, and participants in the International Internet Engineering Task Force (IETF), Daniel devoted to this exciting In the new field. His W3C standard is currently widely used in most automatic interactive voice response (IVR) systems. Due to its outstanding contributions in the field of automatic speech recognition, Daniel has won two "Speech Technology Magazine" in Daniel (SPEECH Magazine).
Huib is currently a GOOGLE product manager, with rich experience in the browser industry, and is currently leading the team to engage in the research and development of WebRTC 1.0 in Chrome. Before joining Google, he has been in the Opera leadership team team. He has made great contributions to browser experience innovation and integrates WebRTC with the engineer team in Opera. In Sweden, Huib and other engineers of Google conducted research and development of WEBRTC projects. A multi-touch invented invented in the Philips Institute, such as multi-touch due to apple phones.
Master the future trend of RTC technology standards, start here, read full text, original title: From AV1, AVS to WebRTC, they will tell you the future trend of technical standards
Article Source: [Micro Signal: Shengwang-Agora, WeChat public number: Voice Network Agora] Welcome to add attention! Please indicate the source of the article.
Our other product:
