Video processing apparatus and method for allocating addresses to data of macroblocks and storing the same, and medium containing a program for performing such method

Abstract

One macroblock is constructed by 16×16 pixels (bytes). When the macroblocks are stored in a frame memory constructed by a DRAM or the like, addresses are sequentially allocated in an ascending order in a manner such that addresses 0000 to 0255 are allocated to the first macroblock and addresses 0256 to 0512 are allocated to the second macroblock and the macroblocks are stored. The macroblocks stored in this manner are read out in the ascending order of the addresses.

1 . A video processing apparatus for decoding a video stream having a layer structure constructed by a sequence layer, a GOP layer, a picture layer, a slice layer, a macroblock layer, and a block layer, comprising: 
 Sequence_Header information predicting means for predicting Sequence_Header information on the basis of information which certainly appears in a picture; and    decoding means for decoding video data by using the information predicted by said Header information predicting means when said Sequence_Header is not detected.    
   
   
       2 . A video processing apparatus according to  claim 1 , wherein said Sequence_Header information predicting means includes vertical pixel number predicting means for predicting the number of pixels in the vertical direction of a picture plane from information in a slice.  
   
   
       3 . A video processing apparatus according to  claim 1 , wherein said Sequence_Header information predicting means includes horizontal pixel number predicting means for predicting the number of pixels in the horizontal direction of a picture plane from information in a macroblock.  
   
   
       4 . A video processing apparatus according to  claim 1 , wherein said Sequence_Header information predicting means includes aspect ratio predicting means for predicting the number of pixels in the vertical direction of a picture plane from information in a slice, predicting the number of pixels in the horizontal direction of the picture plane from information in a macroblock, and predicting an aspect ratio of the picture plane from said predicted number of pixels in the vertical direction and said predicted number of pixels in the horizontal direction.  
   
   
       5 . A video processing method of decoding a video stream having a layer structure constructed by a sequence layer, a GOP layer, a picture layer, a slice layer, a macroblock layer, and a block layer, comprising the steps of: 
 predicting Sequence Header information on the basis of information which certainly appears in a picture; and    decoding video data by using said predicted information when said Sequence_Header is not detected.    
   
   
       6 . A video processing method according to  claim 5 , wherein when said Sequence_Header information is predicted, the number of pixels in the vertical direction of a picture plane is predicted from information in a slice.  
   
   
       7 . A video processing method according to  claim 5 , wherein when said Sequence_Header information is predicted, the number of pixels in the horizontal direction of a picture plane is predicted from information in a macroblock.  
   
   
       8 . A video processing apparatus according to  claim 5 , wherein when said Sequence Header information is predicted, the number of pixels in the vertical direction of a picture plane is predicted from information in a slice, the number of pixels in the horizontal direction of the picture plane is predicted from said number of pixels in the vertical direction and said number of pixels in the horizontal direction and an aspect ratio of the picture plane is predicted.