Video/audio signal processing method and video-audio signal processing apparatus
First Claim
Patent Images
1. A video/audio signal processing method for processing supplied compression-encoded video/audio signals, said method comprising the steps of:
- parsing said video/audio signals in a compressed domain of the video/audio signals and extracting therefrom motion vectors of said video/audio signals, DCT-coefficients and macroblock-type;
using said extracted motion vectors, DCT-coefficients and macroblock-type to extract at least one compressed domain feature point representing characteristics of said video/audio signals in a compressed domain of said video/audio signals;
performing motion estimation of the extracted feature points;
tracking the feature points associated with a motion vector through a pre-set number of frames of said video/audio signals; and
calculating and extracting the block signature for the current block of high relevance as selected in a discrete-cosine-transform domain using part or all of DCT-coefficients in a block,wherein said extraction step includes a step of calculating the block relevance metric of all blocks according to said DCT-coefficients in the current frame to determine a block having high relevance as a candidate of the feature point selected as the next feature point based on said motion estimation step,wherein said extraction step includes a step of performing inverse transform of transforming said compressed domain only for the blocks of high relevance selected by said metric calculating step and of performing motion compensation for a prediction coded macroblock or a bidirectionally prediction coded macroblock.
1 Assignment
0 Petitions
Accused Products
Abstract
A metadata extraction unit has a feature point selection and motion estimation unit 62 for extracting at least one feature point representing characteristics of the video/audio signals in a compressed domain of the video/audio signals. Thus, reduction of time or cost for processing can be realized and it makes it possible to process effectively.
-
Citations
47 Claims
-
1. A video/audio signal processing method for processing supplied compression-encoded video/audio signals, said method comprising the steps of:
-
parsing said video/audio signals in a compressed domain of the video/audio signals and extracting therefrom motion vectors of said video/audio signals, DCT-coefficients and macroblock-type; using said extracted motion vectors, DCT-coefficients and macroblock-type to extract at least one compressed domain feature point representing characteristics of said video/audio signals in a compressed domain of said video/audio signals; performing motion estimation of the extracted feature points; tracking the feature points associated with a motion vector through a pre-set number of frames of said video/audio signals; and calculating and extracting the block signature for the current block of high relevance as selected in a discrete-cosine-transform domain using part or all of DCT-coefficients in a block, wherein said extraction step includes a step of calculating the block relevance metric of all blocks according to said DCT-coefficients in the current frame to determine a block having high relevance as a candidate of the feature point selected as the next feature point based on said motion estimation step, wherein said extraction step includes a step of performing inverse transform of transforming said compressed domain only for the blocks of high relevance selected by said metric calculating step and of performing motion compensation for a prediction coded macroblock or a bidirectionally prediction coded macroblock. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A video/audio signal processing apparatus for processing supplied compression-encoded video/audio signals, comprising:
-
means for parsing said video/audio signals in a compressed domain of the video/audio signals to extract therefrom motion vectors of said video/audio signals, DCT-coefficients and macroblock-type; extraction means for using said extracted motion vectors, DCT-coefficients and macroblock-type to extract at least one compressed domain feature point representing characteristics of said video/audio signals in a compressed domain of said video/audio signals; means for performing motion estimation of the extracted feature points; means for tracking the feature points associated with a motion vector through a pre-set number of frames of said video/audio signals; and calculating and extraction means for calculating and extracting the block signature for the current block of high relevance as selected in a discrete-cosine-transform domain using part or all of DCT-coefficients in a block, wherein said extraction means calculates the block relevance metric of all blocks according to said DCT-coefficients in the current frame to determine a block having high relevance as a candidate of the feature point selected as the next feature point based on said motion estimation step, wherein said extraction means includes means for performing inverse transform of transforming said compressed domain only for the blocks of high relevance selected by said metric calculating means and of performing motion compensation for a prediction coded macroblock or a bidirectionally prediction coded macroblock. - View Dependent Claims (25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47)
-
Specification