Stereoscopic 3D-video image digital decoding system and method

Abstract

Described herein is a MPEG-2 compatible stereoscopic 3D-video image digital decoding method and system. In order to obtain 3D-images from a digital video stream, modifications are made to the current MPEG2 decoders, by means of software and hardware changes in different parts of the decoding process. Namely, the video_sequence structures of the video data stream are modified via software to include the necessary flags at the bit level of the image type in the TDVision® technology. Modifications are also made in the decoding processes as well as in decoding the information via software and hardware, wherein a double output buffer is activated, a parallel and difference decoding selector is activated, the decompression process is executed, the corresponding output buffer is displayed; and the decoder is programmed via software to simultaneously receive and decode two independent program streams, each with an TDVision® stereoscopic identifier.

1 . A stereoscopic 3D-video image digital decoding method, in which the structures of the video_sequence of the video data stream are modified via software, to include flags at the bit level for the image type, comprising: 
 modifying the software and by using the user_data( ) section to store the error correction which allows regeneration of the stereoscopic video signal, thereby actually identifying the video format;    applying a logical “and” for MPEG2 backward compatibility in case it is not a TDVision® video;    decoding by scanning the video_sequence; when the image is a TDVision® image: 
 a) storing the last complete image buffer in the left or right channel buffer.  
 b) applying the differences or parallel decoding for B type frame information,  
 c) applying error correction to the last image obtained by applying the motion and color correction vectors,  
 d) storing the results in their respective channel buffer, and  
 e) continuing with the video_sequence reading.  
   
   
   
       2 . The stereoscopic 3D-video image digital decoding method and system, in which the video_sequence structures of the video data stream are modified via software to include the necessary flags at the bit level of the image type of  claim 1 , wherein the decoder compilation format comprises: 
 a) reading video_sequence,    b) discriminating the sequence_header, if a TDVision® image is identified, then activating the double buffer,    c) reading in the user_data the image as if it was contained in said structure,    d) adding in the sequence_scalable_extension information to the video_sequence MPEG, said information could be contained within said structure,    e) finding in the picture_header the TDVision® image identifier in the extra_bit_picture,    f) reading the “B” type image in the picture_coding_extension, and if it is a TDVision® type image, decoding then the second buffer, and    g) if the image is temporarily scalable, applying “B” to the decoder.    
   
   
       3 . The stereoscopic 3D-video images digital decoding method, in which the structures and the video_sequence of the video data stream are modified to include the necessary flags at the bit level of the image type of  claim 1 , wherein when the decoder detects a user_data( ) code, it searches the 32-bit 3DVision®_start_identifier=0x000ABCD identifier, upon detecting this information a call is made to the special decoding function which compares the output buffer and applies it from the current reading offset of the video_sequence.  
   
   
       4 . The stereoscopic 3D-video images digital decoding method, in which the video_sequence structures of the video data stream are modified via software to include the necessary flags at the bit level of the image type of  claim 1 , wherein the decoder is programmed via software to simultaneously receive and decode two program streams.  
   
   
       5 . The stereoscopic 3D-video images digital decoding method, in which the video_sequence structures of the video data stream are modified via software to include the necessary flags at the bit level of the image type of  claim 1 , wherein two interdependent video signals can be sent within the same video_sequence; said signals depending one form the other, and coming from a 3DVision® camera; in terms of their algebraic addition (R−L=delta), each signal is stored as a B type frame, which decoding is by differences from one of them.  
   
   
       6 . The Stereoscopic 3D-video images digital decoding method, in which the video_sequence structures of the video data stream are modified via software to include the necessary flags at the bit level of the image type of  claim 1 , wherein two independent video streams L and R are stored in simultaneous form, but being synchronized with the same time_code, and decoded and displayed in parallel.  
   
   
       7 . A stereoscopic 3D-video image digital decoding system, in which the video_sequence structures of the video data stream are modified via hardware, wherein the specific use of the structures, substructures and sequences belong to the video_sequence to implement the MPEG2 backward-compatible TDVision® technology via hardware comprise: 
 discriminating whether it is a 2S or 3D signal;    activating a double output buffer (additional memory);    activating a parallel decoding selector, activating a difference-decoding selector;    executing the image decompression process, displaying the image in the corresponding output buffer; and    enabling the PICTURE_DATA3D( ) function, which is transparent for the compatible MPEG2 readers.    
   
   
       8 . The stereoscopic 3D-video image digital decoding system of  claim 7 , wherein the specific use of the structures, substructures and sequences belonging to the video_sequence in order to implement the MPEG2 backward-compatible TDVision® technology via hardware comprise: 
 a) sequence_header    aspect_ratio_information    1001 n/a in TDVision®   1010 4:3 in TDVision®   1011 16:9 in TDVision®   1100 2.21:1 in TDVision®   a logical “and” with 0111 is executed to obtain backward compatibility with 2D systems, where an instruction is sent to the Digital Signal Processor (DSP) stating that the stereoscopic pair buffer (left or right) should be equal to the source;    b) frame_rate_code    1001 24,000/1001 (23.976) in TDVision® format    1010 24 in TDVision® format    1011 25 in TDVision® format    1100 30,000/1001 (29.97) in TDVision® format    1101 30 in TDVision® format    1110 50 in TDVision® format    1111 60,000/1001 (59.94) in TDVision® format    a logical “and” with 0111 is executed to obtain backward compatibility with 2D systems, where an instruction is sent to the DSP stating that the stereoscopic pair buffer (left or right) should be equal to the source;    c) user_data( )    sequence_scalable_extension    d) picture_header    extra_bit_picture    0=TDVision®   1=normal    e) picture_coding_extension    picture_structure    00=image in TDVision® format    f) picture_temporal_scalable_extension( ).    
   
   
       9 . The stereoscopic 3D-video images digital decoding system of  claim 7 , wherein when the PICTURE_DATA3D( )structure is recognized, it proceeds to read the information directly by the decoder, but it writes the information in a second output buffer also connected to a video output additional to that existing in the electronic display device.  
   
   
       10 . The stereoscopic 3D-video images digital decoding system of  claim 7 , wherein, if the signal is of TDVision® type, it is identified if it is a transport stream, program stream or left or right multiplexion at 60 frames per second; when it is a transport stream it has backward compatibility in the current 2D coders; where an instruction is sent to the DSP stating that the stereoscopic pair buffer (left or right) should be equal to the source, having the ability to display the video without 3D characteristics of TDVision®.