US 11,943,489 B2
Method and system for automatic real-time frame segmentation of high resolution video streams into constituent features and modifications of features in each frame to simultaneously create multiple different linear views from same video source
Ela Kliots Shapira, Mattan (IL); and Abraham Zvi Yehiel Elbaum, Petah-Tikva (IL)
Assigned to Snakeview Data Science, Ltd., Rishon le Zion (IL)
Filed by Ela Kliots Shapira, Mattan (IL); and Abraham Zvi Yehiel Elbaum, Petah-Tikva (IL)
Filed on Aug. 3, 2022, as Appl. No. 17/817,202.
Application 17/817,202 is a continuation of application No. 16/973,739, granted, now 11,445,227, previously published as PCT/IL2019/050141, filed on Feb. 6, 2019.
Claims priority of provisional application 62/683,654, filed on Jun. 12, 2018.
Prior Publication US 2022/0377395 A1, Nov. 24, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. H04N 21/2343 (2011.01); G11B 27/036 (2006.01); H04N 9/64 (2023.01); H04N 11/24 (2006.01); H04N 21/236 (2011.01); H04N 21/81 (2011.01)
CPC H04N 21/23439 (2013.01) [G11B 27/036 (2013.01); H04N 9/64 (2013.01); H04N 11/24 (2019.01); H04N 21/236 (2013.01); H04N 21/8146 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of automatically modifying high resolution video data in real time using a computer including a GPU, the method comprising:
receiving a video data stream by the computer, wherein the video data stream is a high definition stream modified on a frame-by-frame basis, and includes audio data;
extracting color data for individual pixels of the frames of the video data stream to create color layers;
cascading the color layers into a unified color map of features in the frame;
assembling instructions to modify pixels of the frame at a level of the unified color map of features from stored data according to predefined instructions, wherein the instructions at least include additional graphical data to be composited with original graphical data of the video data stream;
creating multiple modified sets of video data streams according to the instructions, wherein the modified sets of video data are different from each other and the creating includes:
merging the color layers and the additional graphical data into single-layer color frames, wherein each single-layer color frame is based on a discrete set of the instructions,
rendering each single-layer color frame based on output resolution requirements to create at least one final frame,
encoding, based on the resolution, each final frame to at least one codec to create video components, and
multiplexing the video components with the audio data into desired file formats;
and making the modified set of video data streams available for access by one or more intended recipients.