System and method for video encoding using constructed reference frame
First Claim
Patent Images
1. A method of encoding a series of source video frames to be transmitted to a decoder to produce a series of decoded video frames for viewing, the method comprising:
- creating a constructed reference frame using an encoder,wherein the constructed reference frame is a frame in addition to the series of source video frames;
encoding the constructed reference frameto obtain an encoded constructed reference frame;
compressing the series of source video framesto obtain a compressed digital video signal,at least two of the series of source video frames compressed using the constructed reference frame; and
generating a bitstreamincludingthe encoded constructed reference frame andthe compressed digital video signalfor transmission to the decoderfor a subsequent decoding processwherein the constructed reference frame is omittedfrom the series of decoded video framesduring a subsequent viewing process,wherein the series of decoded video frames was obtained from the compressed digital video signal.
4 Assignments
0 Petitions
Accused Products
Abstract
Disclosed herein is a method for digital video encoding prediction comprising creating a constructed reference frame using an encoder and compressing a series of source video frames using the constructed reference frame to obtain a bitstream including a compressed digital video signal for a subsequent decoding process. The constructed reference frame is omitted from the series of digital video frames during the subsequent viewing process.
44 Citations
20 Claims
-
1. A method of encoding a series of source video frames to be transmitted to a decoder to produce a series of decoded video frames for viewing, the method comprising:
-
creating a constructed reference frame using an encoder, wherein the constructed reference frame is a frame in addition to the series of source video frames; encoding the constructed reference frame to obtain an encoded constructed reference frame; compressing the series of source video frames to obtain a compressed digital video signal, at least two of the series of source video frames compressed using the constructed reference frame; and generating a bitstream including the encoded constructed reference frame and the compressed digital video signal for transmission to the decoder for a subsequent decoding process wherein the constructed reference frame is omitted from the series of decoded video frames during a subsequent viewing process, wherein the series of decoded video frames was obtained from the compressed digital video signal. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A method of encoding a series of source video frames to be transmitted to a decoder to produce a series of decoded video frames for viewing, the method comprising:
-
A) encoding each block of a frame of the series of source video frames using an intra mode prediction; B) encoding each block of the frame of the series of source video frames using an inter mode prediction using a motion vector and a reconstruction buffer of a previous frame in the series of source video frames; C) calculating a first reconstructed error score of each block of the frame encoded using the intra mode prediction; D) calculating a second reconstructed error score of each block of the frame encoded using the inter mode prediction; E) calculating a cumulative best score of the frame by summing a lower of either the first reconstructed error score or the second reconstructed error score of each block of the frame; F) summing the first reconstructed error score of each block of the frame to create a cumulative intra error score; G) calculating a percentage of blocks of the frame where the second reconstructed error score is less than the first reconstructed error score; H) calculating a motion vector for each block of the frame; I) calculating an average motion vector of blocks of the frame using the calculated motion vectors; J) performing steps A) through I) for each frame of the series of source video frames; K) creating a first constructed reference frame using an encoder wherein the constructed reference frame is a frame in addition to the series of source video frames; L) defining a minimum number and a maximum number of frames in the series of source video frames that can be encoded using a single constructed reference frame; M) selecting a frame N of the series of source video frames; N) comparing the percentage of blocks of a next frame N+1 subsequent to the frame N where the second reconstructed error score is less than the first reconstructed error score to a value based on the average motion vector of blocks of the frame N; O) incrementing a boost score for the first constructed reference frame using the cumulative intra error score, and a cumulative inter error score and a lesser of the percentage of blocks of the next frame N+1 where the second reconstructed error score is less than the first reconstructed error score and the value based on the average motion vector of blocks of the frame N; P) performing steps N) and O) for at least one additional frame in the series of source video frames that is subsequent to the frame N in place of the frame N; Q) calculating a bit allocation for the first constructed reference frame using the boost score; R) encoding the first constructed reference frame into a bitstream for transmission to the decoder using the bit allocation for the first constructed reference frame; S) compressing at least the minimum number of frames and up to the maximum number of frames in the series of source video frames starting at the frame N using the first constructed reference frame to obtain a compressed digital video signal for inclusion in the bitstream for a subsequent decoding process wherein the first constructed reference frame is omitted from the series of decoded video frames during a subsequent viewing process, the series of decoded video frames obtained from the compressed digital video signal. - View Dependent Claims (17, 18, 19)
-
-
20. A method of viewing a series of decoded video frames from a compressed digital video signal obtained by compressing a series of source video frames, the method comprising:
-
decoding a first constructed reference frame from a bitstream including the compressed digital video signal wherein the first constructed reference frame is a frame in addition to the series of source video frames; using the first constructed reference frame to decode a first set of at least two sequential frames of the series of decoded video frames from a portion of the compressed digital video signal in the bitstream; decoding a second constructed reference frame present in the bitstream after the portion of the compressed digital video signal wherein the second constructed reference frame is a frame in addition to the series of source video frames; using the second constructed reference frame to decode a second set of at least two sequential frames of the series of decoded video frames from another portion of the compressed digital video signal in the bitstream; and arranging the first set and the second set in sequential order for viewing, wherein the first constructed reference frame and the second constructed reference frame are omitted from the sequential order.
-
Specification