Video encoding apparatus and method and video encoding mode converting apparatus and method

Abstract

A video encoding apparatus comprises a motion vector detector to detect a motion vector of an input picture referring to a reference picture, and a predictive encoder to perform forward predictive encoding and bidirectional predictive encoding using the motion vector and the reference picture, the forward predictive encoder subjecting the macroblock of the forward predictive encoded picture to a variable length encoding in not_coded mode when a correlation between the macroblocks of the forward and bidirectional predictive encoded pictures and the reference picture which are located at the same position is high.

1 . A video encoding apparatus to encode a video, comprising: 
 an input picture module configured to receive an input picture;    a memory to store a reference picture;    a motion vector detection module configured to detect a motion vector of the input picture in units of a macroblock by block matching referring to the reference picture; and    a predictive encoding module configured to perform forward predictive encoding and bi-directional predictive encoding in units of a macroblock using the motion vector and the reference picture, to generate a forward predictive encoded picture and at least one bi-directional predictive encoded picture which is inputted between the forward predictive encoded picture and the reference picture,    wherein the predictive encoding module skips a macroblock of the forward predictive encoded picture when correlations among the macroblock of the forward predictive encoded pictures, a macroblock of the bi-directional predictive encoded picture and a macroblock of the reference picture which are located at the same position as that of the forward predictive encoded picture are high.    
   
   
       2 . The video encoding apparatus according to  claim 1 , which includes an orthogonal transformation module configured to subject each macroblock to orthogonal transformation and quantization to generate quantized orthogonal transformed coefficients, and wherein the predictive encoding module skips macroblocks of the forward predictive encoded picture that satisfy following conditions: 
 (a) forward motion vectors of the macroblocks of all bidirectional predictive encoded pictures using the reference picture which is the same as that used for the forward predictive encoded picture are 0, and    (b) matching residuals with respect to the macroblocks of all the bidirectional predictive encoded pictures that are at the same position are not more than a threshold,    
   
   
       3 . The video encoding apparatus according to  claim 2 , wherein the predictive encoding module performs a forward motion vector detection of each of the macroblocks of all bidirectional predictive encoded pictures subjected to the forward motion vector detection by using the reference picture, before a variable length encoding of the macroblock of the forward predictive encoded picture that is located at the same position as those of the bidirectional predictive encoded pictures.  
   
   
       4 . The video encoding apparatus according to  claim 1 , wherein the motion detection module carries out a forward motion vector detection of the macroblocks of all bidirectional predictive encoded pictures that are located at the same position as that of the reference picture before variable length encoding of the macroblock of the forward predictive encoded picture, and the encoding module includes a variable length encoder to subject each macroblock of each of the pictures to variable length encoding.  
   
   
       5 - 8 . (canceled)  
   
   
       9 . A video encoding method of subjecting a video to intraframe encoding, forward predictive encoding, and bidirectional predictive encoding, comprising: 
 detecting a motion vector of an input picture in units of a macroblock by block matching referring to a reference picture;    performing the forward predictive encoding and bi-directional predictive encoding in units of a macroblock using the motion vector and the reference picture, to generate a forward predictive encoded picture and at least one bidirectional predictive encoded picture which is inputted between the forward predictive encoded picture and the reference picture; and    skipping a macroblock of the forward predictive encoded picture when correlations among the macroblock of the forward predictive encoded picture, a macroblock of the bi-directional predictive encoded picture and a macroblock of the reference picture which are located at the same position as that of the forward predictive encoded picture is are high.    
   
   
       10 . A video encoding method of encoding an input video signal, the method comprising: 
 detecting a forward motion vector every macroblock of a to-be-encoded picture using as a reference picture an intraframe encoded picture or a first forward encoded picture just before the to-be-encoded picture to generate a second forward predictive encoded picture and bi-directional predictive encoded pictures;    subjecting pixel values of the macroblock of the to-be-encoded picture to an orthogonal transformation every macroblock and quantizing orthogonal-transformed coefficients;    skipping the macroblock of each of the forward predictive encoded picture and the bi-directional predictive encoded pictures when satisfying following conditions:    (a) a forward motion vector of the macroblock of the bi-directional predictive encoded picture that is located at the same position as the macroblock of the to-be-encoded picture is 0, the bidirectional predictive encoded picture subjected to a motion vector detection using the same reference picture as that used for the to-be-encoded picture, and    (b) differences between macroblocks of all the bidirectional predictive encoded pictures as that of the reference picture and the macroblock of the reference picture are not more than a threshold.    
   
   
       11 - 41 . (canceled)  
   
   
       42 . The video encoding method according to  claim 9 , wherein the performing the forward predictive encoding includes performing a forward motion vector detection of each of the macroblocks of all bidirectional predictive encoded pictures subjected to the forward motion vector detection by using the reference picture, before a variable length encoding of the macroblock of the forward predictive encoded picture that is located at the same position as those of the bidirectional predictive encoded pictures.  
   
   
       43 . The video encoding method according to  claim 9 , wherein the detecting the motion detection includes performing a forward motion vector detection of the macroblocks of all bidirectional predictive encoded pictures that are located at the same position as that of the reference picture before variable length encoding of the macroblock of the forward predictive encoded picture, and the performing the forward predictive encoding includes subjecting each macroblock of each of the pictures to variable length encoding.