公开(公告)号：JP2000165884A

公开(公告)日：2000-06-16

申请号：JP31914499

申请日：1999-11-10

Applicant: ST MICROELECTRONICS SRL

Inventor： PICCINELLI EMILIANO ,  PEZZONI LUCA ,  PAU DANILO

IPC: H04N7/32 ,  G06T9/00 ,  H04N7/26 ,  H04N7/50

Abstract: PROBLEM TO BE SOLVED: To code video data with a more simplified hardware by calculating the complexity index of a single block at present and adding the maximum absolute difference of all blocks for fixing all the complexity indexes of a relative macro block. SOLUTION: First of all, the complexity index of a single block composing of the current macro block is calculated concerning both the styles of 'field' and 'frame'. By adding the maximum value of differences calculated to all the blocks composing of the macro block, entire macro block complexity can be provided. This calculation is repeated twice, namely, concerning the block provided when successive discrete cosine transformations(DCT) are performed in the 'frame' mode, the calculation is performed once on the line of a block corresponding to different semi-fields and the DCT is performed in the 'field' mode. Next, the second calculation is executed.

公开(公告)号：JP2003179925A

公开(公告)日：2003-06-27

申请号：JP2002275956

申请日：2002-09-20

Applicant: ST MICROELECTRONICS SRL

Inventor： BARTOLUCCI VILIM ,  PAU DANILO ,  EMILIANO PICCINELLI

IPC: H04N7/30 ,  G06T9/00 ,  H03M7/30 ,  H04N7/50 ,  H04N19/94

Abstract: PROBLEM TO BE SOLVED: To provide a method for compressing a signal at a higher level compression while maintaining the same quality level and main features of MPEG2 standards. SOLUTION: The process for compressing a digital video signal comprises the steps of starting a frame for constituting the digital video signal from a microblock for receiving movement compensations (13, 14), dividing the frame into the blocks, assembling the set of the coefficient generated by using a discrete cosine transfer (15) so as to generate each set of the coefficients in each block, constituting them to the set of the vectors by masking (M 1 , M 2 , M 3 and M 4 ), detecting the dispersion of the vector, digitizing the vector for usable multiple bits by a pyramid vector digitizer (22), linking the vector to individual digitized pyramids having decided size in response to the number of usable bits, and encoding the digitized vector by corresponding code word. COPYRIGHT: (C)2003,JPO

公开(公告)号：JPH11251919A

公开(公告)日：1999-09-17

申请号：JP33213498

申请日：1998-11-24

Applicant: ST MICROELECTRONICS SRL

Inventor： PAU DANILO ,  BRUNI ROBERTA

IPC: G06T9/00 ,  H03M7/30 ,  H04N19/50 ,  H04N7/32

Abstract: PROBLEM TO BE SOLVED: To reduce a required memory capacity by providing a more efficient method for selecting a quantizer. SOLUTION: This memory required amount reduction method is for reducing a memory required amount for temporarily storing the stream of a digital data block in the encoding/decoding system of information capable of being transferred by a block unit. Data are compressed, encoded and stored in a memory by the block unit by tree search vector quantization(TSVQ), and in a read stage, the encoded and compressed data are decoded and extracted and the stream of the digital data block is reconstituted.

公开(公告)号：JP2003163932A

公开(公告)日：2003-06-06

申请号：JP2002261167

申请日：2002-09-06

Applicant: ST MICROELECTRONICS SRL

Inventor： CONCION DAVIDE ,  EMILIANO PICCINELLI ,  PAU DANILO

IPC: H04N7/30 ,  G06T9/00 ,  H03M7/30 ,  H04N7/26 ,  H04N7/46

Abstract: PROBLEM TO BE SOLVED: To provide format conversion that is achieved by operation in a DCT region. SOLUTION: A DCT macroblock (MB) contained in an MPEG bit stream is divided into a plurality of microblocks (B1, B2, B3, and B4) each. After an important frequency component in each of DCT blocks (B1, B2, B3, and B4) is identified and stored, microblocks (mB1, mB2, mB3, and mB4) at the corresponding upper left portion are separated, and the coefficients of other microblocks are set to zero. Inverse discrete cosine conversion (16) is carried out to the separated microblocks (nB1, mB2, mB3, and mB4), and a merged block is generated by merging with the microblock obtained from other DCT blocks contained in each macroblock (MB). Discrete cosine conversion (20) is performed to the obtained merged block, and assembly is carried out to the macroblock (MBD) of the converted format for obtaining a final block. COPYRIGHT: (C)2003,JPO

公开(公告)号：JP2001119703A

公开(公告)日：2001-04-27

申请号：JP2000273285

申请日：2000-09-08

Applicant: ST MICROELECTRONICS SRL

Inventor： PAU DANILO ,  CHIMIENTI ANTONIO

IPC: H04N11/04 ,  G06T9/00 ,  H03M7/40 ,  H04N7/26 ,  H04N7/32 ,  H04N7/50

Abstract: PROBLEM TO BE SOLVED: To provide a technology that decreases the video memory conditions of a decoder for a compressed video picture sequence. SOLUTION: Re-compressing a picture used for prediction reference purpose after its expansion and before being stored in an external memory can remarkably reduce the capacity of a memory required for decoding.

公开(公告)号：JP2000083257A

公开(公告)日：2000-03-21

申请号：JP22300999

申请日：1999-08-05

Applicant: ST MICROELECTRONICS SRL

Inventor： PAU DANILO ,  BAGNI DANIELE ,  PEZZONI LUCA

IPC: H04N7/32 ,  G06F17/30 ,  H04N5/14 ,  H04N7/26 ,  H04N7/36 ,  H04N7/46 ,  H04N7/50

Abstract: PROBLEM TO BE SOLVED: To obtain the technology for enhancing estimation of a motion in the case of processing a video image for a video compressor and a coder. SOLUTION: In compression and coding of digital data relating to a video sequence consisting of pictures including motion estimation to eliminate temporal redundancy, remarkable advantages of the capability to recognize occurrence of scene changes are obtained, since only estimation of forward motion is used for pictures preceding a scene change and only estimation of backward motion is used for pictures in succession to the scene change of a given sequence to make prediction calculation of pictures. According to this method, a scene change is detected with reliability by checking two separate indices. Thus, accidental detection in the case of pictures with noise, zooming, or other situations that may possibly cause an adverse effect on one or the other index is prevented.

公开(公告)号：JPH11225334A

公开(公告)日：1999-08-17

申请号：JP32157498

申请日：1998-11-12

Applicant: ST MICROELECTRONICS SRL

Inventor： PAU DANILO ,  ROVATI FABRIZIO ,  VALSASNA ANNA ,  BRUNI ROBERTA

IPC: H04N5/93 ,  G06T9/00 ,  H04N7/26 ,  H04N7/50 ,  H04N7/24

Abstract: PROBLEM TO BE SOLVED: To obtain a device which calculates the dispersion value of a macroblock of a digital video by outputting the square value of an image that is continuously inputted, inputting each pixel of the macroblock, counting up, resetting a counter simultaneously with the input of a final pixel, transmitting the pixel value and square value of a pixel to an accumulator and storing accumulative result value of the accumulator. SOLUTION: For an algorithm that is executed by an accelerator for dispersion value calculation, a simple counter LC is used and two demultiplexers transmit an input line which processes in a method such as decides to which dispersion value a prescribed input pixel belongs to eight separate dispersion value operation paths. Relating to a prescribed line pattern in which a macroblock is scanned, as a rule, the bit of the counter LC that drives multiplexers is selected so that a dispersion value can be accurately operated. The counter LC is reset at the time of starting each macroblock and is counted up by the input of each pixel.

公开(公告)号：JP2003208306A

公开(公告)日：2003-07-25

申请号：JP2002347918

申请日：2002-11-29

Applicant: ST MICROELECTRONICS SRL

Inventor： CREMONESI ALESSANDRO ,  ROVATI FABRIZIO ,  PAU DANILO

IPC: G06F9/38 ,  G06F9/30 ,  G06F9/318 ,  G06F9/45

Abstract: PROBLEM TO BE SOLVED: To provide a processing architecture capable of executing two or more different sets of instructions. SOLUTION: The architecture comprises a single CPU configured for executing both the instructions of the first set (OsTask1.1, OsTask1.2,...) and the instructions of the second set (MmTask2.1, MmTask2.2, MmTask2.3,...). The single CPU is configured for being switched between a first operating mode, in which the single CPU executes the first set instructions (OsTask1.1, OsTask1.2,...) and a second operating mode, in which the single CPU executes the second set of instructions (MmTask2.1, MmTask2.2, MmTask2.3,...). The solution can be generalized by the use of a plurality of switching instructions between more than two execution modes for different CPUs. COPYRIGHT: (C)2003,JPO

公开(公告)号：JP2001189662A

公开(公告)日：2001-07-10

申请号：JP2000349015

申请日：2000-11-16

Applicant: ST MICROELECTRONICS SRL

Inventor： PAU DANILO ,  BORTOT LUCA ,  SACCHI MARIA LUISA

IPC: G06T9/00 ,  H03M7/30 ,  H03M7/40 ,  H04N7/26

Abstract: PROBLEM TO BE SOLVED: To provide technology for easily changing the bit rate of the data stream of a video picture. SOLUTION: An input bit stream is divided into the sequence of encoded data and the sequence of a control bit. The output sequence of the control bit is generated by correcting the sequence of the control bit by the function of a different bit rate in a desired output bit stream. The intermediate sequence of data is generated by decoding the sequence of the code data. The output sequence of the encoded data is generated by quantizing data in a previously established step and encoding the intermediate data of these data. These output sequences are joined and the output stream having the desired bit rate is generated.

公开(公告)号：JP2001119698A

公开(公告)日：2001-04-27

申请号：JP2000273887

申请日：2000-09-08

Applicant: ST MICROELECTRONICS SRL

Inventor： PAU DANILO ,  BORTOT LUCA ,  SACCHI MARIA LUISA

IPC: H03M7/30 ,  H04N7/26 ,  H04N7/30 ,  H04N7/50

Abstract: PROBLEM TO BE SOLVED: To reduce complexity on calculation in the case of calculating the action of the macroblock of a video picture. SOLUTION: A predictive error calculation is performed in a spatial frequency domain obtained by discrete cosine transform instead of performing a calculation about the macro block arranged on a source picture, in the case of calculating the action of the macrobloek of the video picture.