Patents by Inventor Dave Zachariah

Dave Zachariah has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 10291917
    Abstract: Implementations of independent temporally concurrent video stream coding may include encoding a plurality of input frames from an input video sequence, wherein the plurality of input frames includes a first input frame. Encoding the plurality of input frames may include generating a first plurality of encoded frames based on the plurality of input frames such that the first plurality of encoded frames includes a first encoded I-frame corresponding to the first input frame, and generating a second plurality of encoded frames based on the plurality of input frames such that the second plurality of encoded frames includes a first encoded P-frame corresponding to the first input frame. Implementations of independent temporally concurrent video stream coding may include including the first plurality of encoded frames and the second plurality of encoded frames in an output, and transmitting the output to a decoder.
    Type: Grant
    Filed: August 25, 2015
    Date of Patent: May 14, 2019
    Assignee: Google LLC
    Inventors: Ermin Kozica, Dave Zachariah, Willem Bastiaan Kleijn
  • Publication number: 20160065967
    Abstract: Implementations of independent temporally concurrent video stream coding may include encoding a plurality of input frames from an input video sequence, wherein the plurality of input frames includes a first input frame. Encoding the plurality of input frames may include generating a first plurality of encoded frames based on the plurality of input frames such that the first plurality of encoded frames includes a first encoded I-frame corresponding to the first input frame, and generating a second plurality of encoded frames based on the plurality of input frames such that the second plurality of encoded frames includes a first encoded P-frame corresponding to the first input frame. Implementations of independent temporally concurrent video stream coding may include including the first plurality of encoded frames and the second plurality of encoded frames in an output, and transmitting the output to a decoder.
    Type: Application
    Filed: August 25, 2015
    Publication date: March 3, 2016
    Inventors: Ermin Kozica, Dave Zachariah, Willem Bastiaan Kleijn
  • Patent number: 9137561
    Abstract: Implementations of independent temporally concurrent video stream coding may include generating a sequence of encoded frames by encoding a plurality of input frames, wherein encoding the plurality of input frames may include generating a first plurality of encoded frames based on the plurality of input frames, the first plurality of encoded frames including a first plurality of intra-coded frames and a first plurality of inter-coded frames, and independently generating a second plurality of encoded frames based on the plurality of input frames, wherein the second plurality of encoded frames includes a second plurality of intra-coded frames and a second plurality of inter-coded frames, such that the first plurality of encoded frames and the second plurality of encoded frames are temporally concurrent, and such that the intra-coded frames from the second plurality of intra-coded frames are temporally nonconcurrent with the intra-coded frames from the first plurality of intra-coded frame.
    Type: Grant
    Filed: November 12, 2013
    Date of Patent: September 15, 2015
    Assignee: GOOGLE INC.
    Inventors: Ermin Kozica, Dave Zachariah, Willem Bastiaan Kleijn
  • Publication number: 20140079123
    Abstract: Implementations of independent temporally concurrent video stream coding may include generating a sequence of encoded frames by encoding a plurality of input frames, wherein encoding the plurality of input frames may include generating a first plurality of encoded frames based on the plurality of input frames, the first plurality of encoded frames including a first plurality of intra-coded frames and a first plurality of inter-coded frames, and independently generating a second plurality of encoded frames based on the plurality of input frames, wherein the second plurality of encoded frames includes a second plurality of intra-coded frames and a second plurality of inter-coded frames, such that the first plurality of encoded frames and the second plurality of encoded frames are temporally concurrent, and such that the intra-coded frames from the second plurality of intra-coded frames are temporally nonconcurrent with the intra-coded frames from the first plurality of intra-coded frame.
    Type: Application
    Filed: November 12, 2013
    Publication date: March 20, 2014
    Applicant: GOOGLE INC.
    Inventors: Ermin Kozica, Dave Zachariah, Willem Bastiaan Kleijn
  • Patent number: 8582662
    Abstract: The invention relates to methods and apparatuses for encoding and decoding of a video sequence. In connection with encoding/decoding a video sequence it is desirable to increase the video quality without having to increase the bit-rate for the encoded video too much, thereby still providing a bit-efficient representation of the video. If multiple descriptions of the video sequence is used the invention improves the video quality without any increase of the bit-rate. According to the invention, this is achieved by using two or more coding units for encoding the same video sequence, wherein the encoding units perform their encoding operations displaced in time in relation to each other. Correspondingly, two or more decoding units are used for decoding the same video sequence, wherein the decoding units perform their decoding operations displaced in time in relation to each other.
    Type: Grant
    Filed: October 25, 2011
    Date of Patent: November 12, 2013
    Assignee: Google Inc.
    Inventors: Ermin Kozica, Dave Zachariah, Willem Bastiaan Kleijn
  • Publication number: 20120039392
    Abstract: The invention relates to methods and apparatuses for encoding and decoding of a video sequence. In connection with encoding/decoding a video sequence it is desirable to increase the video quality without having to increase the bit-rate for the encoded video too much, thereby still providing a bit-efficient representation of the video. If multiple descriptions of the video sequence is used the invention improves the video quality without any increase of the bit-rate. According to the invention, this is achieved by using two or more coding units for encoding the same video sequence, wherein the encoding units perform their encoding operations displaced in time in relation to each other. Correspondingly, two or more decoding units are used for decoding the same video sequence, wherein the decoding units perform their decoding operations displaced in time in relation to each other.
    Type: Application
    Filed: October 25, 2011
    Publication date: February 16, 2012
    Applicant: Google Inc.
    Inventors: Ermin Kozica, Dave Zachariah, William Bastiaan Kleijn
  • Patent number: 8073049
    Abstract: The invention relates to methods and apparatuses for encoding and decoding of a video sequence. In connection with encoding/decoding a video sequence it is desirable to increase the video quality without having to increase the bit-rate for the encoded video too much, thereby still providing a bit-efficient representation of the video. If multiple descriptions of the video sequence is used the invention improves the video quality without any increase of the bit-rate. According to the invention, this is achieved by using two or more coding units for encoding the same video sequence, wherein the encoding units perform their encoding operations displaced in time in relation to each other. Correspondingly, two or more decoding units are used for decoding the same video sequence, wherein the decoding units perform their decoding operations displaced in time in relation to each other.
    Type: Grant
    Filed: January 31, 2008
    Date of Patent: December 6, 2011
    Assignee: Google Inc.
    Inventors: Ermin Kozica, Dave Zachariah, William Bastiaan Kleijn
  • Publication number: 20080205520
    Abstract: The invention relates to methods and apparatuses for encoding and decoding of a video sequence. In connection with encoding/decoding a video sequence it is desirable to increase the video quality without having to increase the bit-rate for the encoded video too much, thereby still providing a bit-efficient representation of the video. If multiple descriptions of the video sequence is used the invention improves the video quality without any increase of the bit-rate. According to the invention, this is achieved by using two or more coding units for encoding the same video sequence, wherein the encoding units perform their encoding operations displaced in time in relation to each other. Correspondingly, two or more decoding units are used for decoding the same video sequence, wherein the decoding units perform their decoding operations displaced in time in relation to each other.
    Type: Application
    Filed: January 31, 2008
    Publication date: August 28, 2008
    Applicants: GLOBAL IP Solutions, Inc.
    Inventors: Ermin Kozica, Dave Zachariah, Willem Bastiaan Kleijn