Michelle Clancy ©RapidTVNews | 05-04-2012
YouTube has added automatic 3D conversion for short-form videos uploaded in 1080p.
The functionality is a follow-on from a beta feature that let creators convert YouTube videos into 3D with a click, which has resulted in hundreds of thousands of 3D videos that have been shared. Now, the beta has been rolled out to everyone. Users simply select 3D viewing in the quality settings on the YouTube player.
In terms of the underlying conversion technology, YouTube uses a combination of video characteristics such as color, spatial layout and motion to estimate a depth map for each frame of a monoscopic video sequence. Machine learning from the growing number of true 3D videos on YouTube helps the system to learn video depth characteristics and apply them in depth estimation. The generated depth map and the original monoscopic frame create a stereo 3D left-right pair, that a stereo display system needs to display a video as 3D.
Then, YouTube applies cloud computing scalability to break up a video into tiny chunks of data and process them in parallel.
The process was laid out by Deb Mukherjee, technical staff, and Chen Wu, software engineer, in the official YouTube blog.
ضع تعليق باستخدام حساب الفيس بوك
|مواضيع ذات صلة مع YouTube fills out 3D video capability for user-generated content|
|Video youtube New YEAR Celebrations in Amsterdam 2014|
|Video youtube New YEAR Celebrations in New Zealand 2014|
|Video youtube New YEAR Celebrations in Taipei 2014|
|Switzerland-based video-on-demand content aggregator Diva has beta-launched its new Viewster consumer offer.|
|YouTube testing live video streaming|