註冊日期: May 2003
來 自: 台灣'台中
回覆: Stand by for H.266 compression
最初由 草壁梅 發表
February 1, 2018
A seminar organised in London by compression specialist ATEME and audio and video experts from Dolby Labs heard updated news on the planned roll-out of the latest plans for video compression. The event was the latest in a series of industry briefings held in Munich and Paris and another scheduled for Denver.
Jérôme Vieron, (Director, R&D) and Ateme’s acknowledged expert told delegates that a new Codec was well on its way ideally suited for 4K “and even 8K, 12K and 16K…”. He admitted that the acronym-heavy industry bodies and some 4K-branded solutions were frequently confusing for the public and even others involved in the sales and marketing of TVs and related technologies!
Jason Power, Snr Director of marketing/broadcast at Dolby Labs, said that the progress made by Dolby Vision was further boosted when the codec was accepted at CES by Apple for its portfolio of video products, and where Apple was aggressively promoting High Dynamic Range (HDR) as “the new normal” for broadcasters. There were now more than 200 titles available in HDR from Apple. He explained that LGD, Sony, Philips, Hisense, Vestel and others had also adopted Dolby Vision within their latest display ranges, as had certain Games companies and Lenovo laptops.
Vieron told delegates that one new advanced codec, (Future Video Codec/FVC H.266) was now well advanced, and its aim was to achieve a ‘better than 30 per cent compression improvement over existing standards.
FVC/H.266 Development Schedule
• Standardisation process has started
• Target >50 per cent over HEVC
• Oct 2017, Call for Proposals
• Feb 2018, Responses evaluation
• Oct 2018, First test models due
• Oct 2019, First versions of Standard
• End 2020, Final Standard
• June 2021, First hardware Codecs
Vieron added that royalty fees for IP were not always helpful to broadcasters and equipment vendors, not least when requests were made by some to sign NDA’s prior to knowing what the licence fee actually covered or involved.
“But these are new paradigms, and new ways of thinking,” stated Vieron, and one of these was to – in effect – start thinking with a clean sheet of paper and where Ateme was working on a concept where Artificial Intelligence was an integral part of the codec, and actively looked at the content, its target device and where there was “machine deep learning” in the codec. The potential, said Vieron, was between 30-60 percent compression gains in bandwidth. The gain in Sport would be smaller, but the benefit in movie transmission would be at the higher end of the target scale.
向版主報告此篇 | IP位址: 已記錄