Fang Wu

6335990, Jan 1, 2002

Dec 4, 1997

08/985377

Wen H. Chen - Sunnyvale CA
Ji Zhang - San Jose CA
Fang Wu - San Jose CA

Cisco Technology, Inc. - San Jose CA

G06T 510

382261, 382265, 382275, 348607, 348608, 348618, 348620, 37524029

A filter that filters in the spatial and temporal domain in a single step with filtering coefficients that can be varied depending upon the complexity of the video and the motion between the adjacent frames comprises: a IIR filter, a threshold unit, and a coefficient register. The IIR filter and threshold unit are coupled to receive video data. The IIR filter is also coupled to the coefficient register and the threshold unit. The IIR filter receives coefficients, a, from the coefficient register and uses them to filter the video data received. The IIR filter filters the data in the vertical, horizontal and temporal dimensions in a single step. The filtered data output by the IIR filter is sent to the threshold unit. The threshold unit compares the absolute value of the difference between the filtered data and the raw video data to a threshold value from the coefficient register, and then outputs either the raw video data or the filtered data. The present invention is advantageous because it preserves significant edges in video sequence; it preserves motion changes in video sequences; it reduces noise; and it uses minimal memory storage and introduces minimal processing delay.

6445314, Sep 3, 2002

Mar 1, 2000

09/516916

Ji Zhang - San Jose CA
Fang Wu - San Jose CA

Cisco Technology Inc. - San Jose CA

H03M 700

341 67, 341 65

A system for decoding variable length codes comprises a window buffer, a unique variable length code look-up table and a decoder. The window buffer is coupled to receive a bit stream and provides a window output having the same number of bits in the longest variable length code. The output of the window buffer is coupled to address the variable length code look-up table. The look-up table has entries pre-calculated based on the variable length code book and is pre-stored in the system. The window buffer can have a size of any number of bits from X to Y where X is the number of bits in the longest variable length code and Y is a number greater than X. The output of the variable length code look-up table is provided to the decoder. The output of the variable length code look-up table includes the code specified by the bits, and a number of bits that the window buffer should be incremented. This number is in turn used by the decoder to increment the window buffer.

6483543, Nov 19, 2002

Feb 3, 1999

09/244326

Ji Zhang - San Jose CA
Scott Stovall - Bonny Doon CA
Fang Wu - San Jose CA
Yitong Tse - San Jose CA

Cisco Technology, Inc. - San Jose CA

H04N 718

348390, 37524008

A system for transcoding multiple channels of compressed video streams using a self contained data unit such as an autonomous frame includes an autonomous frame processing unit having an autonomous frame generator and an autonomous frame recoder. The autonomous frame generator receives video data and divides it into a series of autonomous frames. Each autonomous frame preferably comprises 1) a frame header including all header information from the original video data plus enough additional information to allow the frame to be recoded using pre-defined autonomous frame structure, and 2) a frame payload including the original video data information. The autonomous frame recoder process the autonomous frames including extracting processing parameters, extracting the video data and setting up or initializing the recoder to process the extracted video data. The autonomous frame recoder preferably further comprises a parser coupled to an initialization unit and a recoder. The present invention also includes a method for processing video data including the steps of: receiving a video bitstream, storing recoding information, dividing the video bitstream into a plurality of autonomous frames each frame including a portion of the video bitstream and recoding information, outputting the plurality of autonomous frames, receiving the plurality of autonomous frames, extracting processing information from the autonomous frame, extracting video data from the autonomous frame, setting the recoding according to the processing information and recoding the extracted video data.

6751259, Jun 15, 2004

Oct 11, 2002

10/270852

Ji Zhang - San Jose CA
Scott Stovall - Bonny Doon CA
Fang Wu - San Jose CA
Yitong Tse - San Jose CA

Cisco Technology, Inc. - San Jose CA

H04N 718

37524026, 37524028

A system for transcoding multiple channels of compressed video streams using a self contained data unit such as an autonomous frame includes an autonomous frame processing unit having an autonomous frame generator and an autonomous frame recoder. The autonomous frame generator receives video data and divides it into a series of autonomous frames. Each autonomous frame preferably comprises 1) a frame header including all header information from the original video data plus enough additional information to allow the frame to be recoded using pre-defined autonomous frame structure, and 2) a frame payload including the original video data information. The autonomous frame recoder process the autonomous frames including extracting processing parameters, extracting the video data and setting up or initializing the recoder to process the extracted video data. The autonomous frame recoder preferably further comprises a parser coupled to an initialization unit and a recoder. The present invention also includes a method for processing video data including the steps of: receiving a video bitstream, storing recoding information, dividing the video bitstream into a plurality of autonomous frames each frame including a portion of the video bitstream and recoding information, outputting the plurality of autonomous frames, receiving the plurality of autonomous frames, extracting processing information from the autonomous frame, extracting video data from the autonomous frame, setting the recoding according to the processing information and recoding the extracted video data.

6891854, May 10, 2005

Dec 19, 2000

09/741451

Ji Zhang - San Jose CA, US
Wen H. Chen - Sunnyvale CA, US
Fang Wu - San Jose CA, US

Cisco Technology, Inc. - San Jose CA

H04J003/16
H04N011/04
H04N007/12

370468, 370538, 37524001, 3483841, 3483901

Digitally compressed video/audio bit streams, when transmitted over digital communication channels such as digital subscriber loop (DSL) access networks, ATM networks, satellite, or wireless digital transmission facilities, can be corrupted due to lack of sufficient channel bandwidth. This invention describes schemes to ensure lossless transmission of bit streams containing pre-compressed video signals within the communication channels. The schemes herein comprises a rate conversion system that converts the bit rate of a pre-compressed video bit stream from one bit rate to another, and that is integrated with a digital communication channel, and to convey the maximum channel transmission rate to the rate conversion system to allow satisfactory transmission of the bit stream from the input of the rate converter through the transmission facility.

7016337, Mar 21, 2006

Feb 28, 2000

09/514577

Fang Wu - San Jose CA, US
Sangeeta Ramakrishnan - San Jose CA, US
Ji Zhang - San Jose CA, US

Cisco Technology, Inc. - San Jose CA

H04L 12/66

370352

Methods and apparatus are provided for statistical re-multiplexing of multiple channels. Mechanisms are provided to manipulate and/or recode multiple compressed bit streams such that a resulting bit stream has a rate matching adjusted based on the allowable output channel rate. Loss-less transmission of compressed video bit streams can be output in real-time. In one example, the system includes encoders and demultiplexers on multiple channels. The demultiplexers are connected to a scheduler and multiplexer to provide an output bit stream at a given channel rate.

7162097, Jan 9, 2007

May 29, 2001

09/867966

Fang Wu - San Jose CA, US
Ke Shen - Sunnyvale CA, US
Yi Tong Tse - San Jose CA, US
Sangeeta Ramakrishnan - San Jose CA, US

Cisco Technology, Inc. - San Jose CA

G06K 9/46

382250

According to the present invention, methods and apparatus for filtering transform coefficients are provided. Transform coefficients associated with an input bit sequence can be zeroed to provide modified transform coefficients associated with an output bit sequence. The transform coefficients may represent information in a portion of data such as a video frame, a picture, or an audio stream. In one example, the transform coefficients are DCT coefficients of an MPEG-2 video sequence. The transform coefficients are filtered to provide modified transform coefficients associated with an output bit sequence. The output bit sequence is a rescaled version of the input bit sequence.

7173947, Feb 6, 2007

Nov 28, 2001

09/998546

Sangeeta Ramakrishnan - San Jose CA, US
Fang Wu - San Jose CA, US
Xiaomei Liu - Milpitas CA, US

Cisco Technology, Inc. - San Jose CA

H04J 3/02
H04J 1/16
H04J 3/14

370537, 370242, 370252

The present invention relates to methods and apparatus for quantitatively measuring the performance of a statistical remultiplexer. The methods of the present invention provide for generation of evaluation metrics associated with at least one of amount of bit rate reduction performed by the statistical remultiplexer, change in video quality attributable to the statistical remultiplexer, wasted output bandwidth by the statistical remultiplexer, decoder buffer level fullness, bit rate reduction characteristics of the statistical remultiplexer, and time delay attributable to the statistical remultiplexer. The methods of the present invention may be implemented by an evaluator that may be incorporated into a statistical remultiplexer or may be separate from and connectable to a statistical remultiplexer.