Color video codec method and system

Abstract

In one aspect of a preferred embodiment, the present invention comprises a system for video compression comprising a video preprocessor; a predictor configured to receive video data from the preprocessor; and an encoder configured to communicate with the predictor. Preferably, the preprocessor comprises a colorspace converter, a frame activity detector, and a subsampler, the predictor comprises a frame differencer and a reference frame handler, and the encoder comprises an error image encoder and an image adder. In another aspect, the invention comprises a system for video decompression comprising a predictor and a decoder configured to communicate with the predictor. Preferably, the predictor comprises a reference frame handler and wherein said decoder comprises an error image decoder and a colorspace converter.

1 . A system for video compression, comprising: 
 a video preprocessor;    a predictor configured to receive video data from said preprocessor; and    an encoder configured to communicate with said predictor;    wherein said preprocessor comprises a colorspace converter, a frame activity detector, and a subsampler;    wherein said predictor comprises a frame differencer and a reference frame handler; and    wherein said encoder comprises an error image encoder and an image adder.    
   
   
       2 . A system as in  claim 1 , wherein said colorspace converter converts from RGB colorspace to YST colorspace.  
   
   
       3 - 4 . (canceled)  
   
   
       5 . A method for video compression, comprising: 
 receiving color video data represented in a first colorspace representation;    converting said received color video data to a second colorspace representation;    identifying activity between consecutive frames of said converted color video data;    subsampling said converted color video data;    calculating error image data based on said subsampled and converted color video data and on said identified frame activity;    encoding said error image data; and    transmitting said encoded error image data to a device capable of displaying color video data;    wherein said step of identifying activity is performed before said step of subsampling.    
   
   
       6 . A method for video decompression, comprising: 
 receiving encoded color video error image data;    decoding said data;    combining said decoded data with previously received data to construct video frame data in a first colorspace representation;    converting said color video frame data to a second colorspace representation with one pass through the data; and    displaying said color video frame data;    wherein said step of converting comprises upsampling and dithering.    
   
   
       7 . A method as in  claim 6 , wherein said step of converting is performed using look-up tables.  
   
   
       8 . A method for representing color video information, comprising: 
 receiving 24-bit RGB color video data; and    transforming said RGB data according to the linear transformation:    Y=18R+36G+6B; S=18R−18B; and T=−18R+36G−18B.    
   
   
       9 . A method for compressing and decompressing color video data, comprising: 
 receiving color video data represented in a first colorspace representation and with a first pixel depth;    converting said color video data to a second colorspace representation with a second pixel depth;    compressing said converted data; and    decompressing said compressed converted data;    wherein said step of decompressing comprises converting said data to a third colorspace representation with a third pixel depth.    
   
   
       10 . A method as in  claim 9 , wherein said second colorspace representation and second pixel depth are selected so as to optimize compression and decompression computational efficiency.  
   
   
       11 . A method as in  claim 9 , wherein said third colorspace representation with a said third pixel depth is selected to comply with format requirements of a display device.  
   
   
       12 . A method as in  claim 9 , wherein said first and third colorspace representations are the same format.  
   
   
       13 . A method as in  claim 9 , wherein said first colorspace representation is RGB.  
   
   
       14 . A method as in  claim 9 , wherein said second colorspace representation is YST.  
   
   
       15 . A method as in  claim 9 , wherein said first pixel depth is 24-bit.  
   
   
       16 . A method as in  claim 9 , wherein said second pixel depth is 12-bit.  
   
   
       17 . A method as in  claim 16 , wherein said third pixel depth is 8-bit, 12-bit, 16-bit, or 24-bit.  
   
   
       18 . A method as in  claim 9 , wherein said third pixel depth is 8-bit, 12-bit, 16-bit, or 24-bit.  
   
   
       19 . A method as in  claim 18 , wherein said third pixel depth is 8-bit.  
   
   
       20 . A method as in  claim 9 , wherein said step of converting said data to a third colorspace representation with a third pixel depth is performed with one pass through the data and comprises upsampling and dithering.