International Business Machines Corporation

Agnes Yee Ngai

Barbara A Hall

Edward Francis Westermann

John Mark Kaczmarczyk

James David Greenfield

Charles E Boice

Michael Patrick Vachon

Robert L Woodard

John A Murdock

Charlene Ann Gebler

John M Sutton

Stephen Philip Pokrinchak

Adrian Stephen Butter

Charles John Stein

Everett G Vail, Iii

Lilla Boroczky

Robert J Yagley

John Michael Sutton

Amy K Jen

Diane Marie Mauersberg

Elliot N Linzer

Bartholomew Blaner

Stamatis Vassiliadis

Cesar Augusto Gonzales

Michael Putrino

Lawrence D Curley

John David Jabusch

Eric Viscito

Alfred T Rundle

Ronald S Svec

Jeffery Dean Carr

Kevin C Huang

Roger Sherman Rutter

Prasoon Tiwari

Daniel J Buerkle

Mark D Atkins

Nader Mohsenian

Ann E Huffman

Patricia A Gage

Cecelia L Dopp

Robert Leslie Woodward

Plural encoders operating in parallel to achieve a desired data rate have their respective outputs combined by an autonomously operating arrangement for transfer of data to a direct memory access arrangement from respective encoders in order in response to a signal asserted upon completion of encoding and output of encoded data corresponding to a predetermined portion of input data. Buffering of encoder output can be either internal or external to the encoders. Zero bytes which may be inherently generated at the beginning and end of an encoder output stream may be suppressed to improve encoded signal quality and efficiency.

Cascaded output for an encoder system using multiple encoders

The present invention provides a system and method for processing a High Definition Television (HDTV) image. Specifically, the system and method of the present invention provides a plurality of programmable encoders connected in parallel and directly (i.e., gluelessly) attached to a HDTV video source. The system and method of the present invention allows a fall HDTV image to be received by each encoder at 74.25 MHz directly from the HDTV video source. Based on programming, each encoder will process only a portion of the full image.

System and method for processing a high definition television (HDTV) image

A distributed control strategy is provided for dynamically encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel. The control strategy is a single pass strategy which allows individual encode bit rates to be dynamically adjusted for each video data stream based in part on relative complexity of the multiple streams of video data, as well as fullness of compressed video data buffers coupled between the encoders and the constant bit rate channel. The control strategy includes encoding the multiple streams of video frames in parallel using multiple encoders, exchanging one or more statistics between the encode processes using an exchange interface, and dynamically adapting encoding of at least one stream of the video frames using the exchanged statistics based on relative complexity of the video frames.

Distributed control strategy for dynamically encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel

A technique is provided for adaptively encoding in hardware, software or a combination thereof a sequence of frames in real time, wherein one or more of the frames includes a random noise portion. The technique includes using statistics analysis to determine whether a current frame includes a random noise portion, and if so, to evaluate and dynamically encode each macroblock thereof based on activity level of the macroblock. Evaluating macroblock activity level includes determining whether its activity level exceeds a predefined threshold indicative of random noise. The macroblock is adaptively encoded by adjusting one or more coding parameters if the macroblock activity level is excessive and its target bitrate is low. For example, when the macroblock is within the random noise portion of the frame, the macroblock is biased towards being coded predictive and an adjusted quantization level is calculated to conserve bits used in encoding the macroblock, thereby moving encode bits from macroblocks within the random noise portion of the frame to macroblocks within the normal portion of the frame.

Adaptively encoding a picture of contrasted complexity having normal video and noisy video portions

A digital video encoder is presented adapted for dynamically switching between sets of quantizer matrix tables without pausing encoding of a stream of video data. Two or more sets of quantizer matrix tables are held at the encoder's quantization unit and compressed store interface for dynamically switching between sets of quant matrix tables at a picture boundary of the sequence of video data, i.e., without stopping encoding of the sequence of video data. Further, while one set of matrix tables is being employed to quantize the stream of video data, the encoder can be updating or modifying another set of quantization matrix tables, again without stopping encoding of the sequence of video data.

Dynamically switching quant matrix tables within an MPEG-2 encoder

A technique is provided for programmably horizontally filtering pixel values of frames of a plurality of video frames. The technique includes, in one embodiment, passing pixel values through a real-time horizontal filter disposed as preprocessing logic of a video encode system. The horizontal filter is programmable and includes a filter coefficients buffer for holding multiple sets of filter coefficients. The horizontal filter programmably employs the multiple sets of filter coefficients to selectively perform spatial noise filtering, or spatial noise filtering and image scaling on the pixels. The filter coefficients are also programmable and may be changed dynamically and repeatedly, with changes being applied at frame boundaries. When performing image scaling, multiple sets of filter coefficients are employed.

Programmable horizontal filter with noise reduction and image scaling for video encoding system

A technique for encoding macroblocks of a frame is provided which includes switching a coding decision for a macroblock from a non-intra decision to a skip macroblock decision when the absolute prediction error for the macroblock is less than a prediction error threshold and the magnitude of the motion vector for the at least one macroblock is less than a motion vector threshold. In one embodiment, the prediction error threshold is multiple programmable prediction error thresholds, which are for comparison with the absolute luminance and chrominance summation components and the maximum luminance and chrominance difference values. The motion vector threshold may be a preset value when the macroblock is to be predictive coded, and may be the magnitude of the motion vector of the previous macroblock when the macroblock is to be bi-directionally encoded.

Macroblock coding technique with biasing towards skip macroblock coding

A technique is provided for programmably vertically filtering pixel values of frames of a sequence of video frames. The technique includes separating luminance components and chrominance components of the pixel values within a vertical filter buffer, then vertically filtering luminance components of the pixel values using programmable luminance filter coefficients, and vertically filtering chrominance components of the pixel values using programmable chrominance filter coefficients. The filtered luminance component data and filtered chrominance component data is subsequently merged onto a single luminance/chrominance bus for further filtering and/or encoding. The luminance and chrominance filter coefficients are programmable and may be changed dynamically and repeatedly at picture boundaries. In one embodiment, the programmable vertical filter includes a four tap luminance component filter and a five tap chrominance component filter.

Programmable vertical filter for video encoding

Method and encoder for encoding a digital video image stream. The encoding includes spatial compression of still images in the video stream and temporal compression between the still images. The spatial compression is carried out by converting a time domain image of a macroblock to a frequency domain image of the macroblock, taking the discrete cosine transform of the frequency domain image, transforming the discrete cosine transformed macroblock image by a quantization factor, and run length encoding the quantized discrete cosine transformed macroblock image. The temporal compression is carried out by reconstructing the run length encoded, quantized, discrete cosine transformed image of the macroblock, searching for a best match macroblock, and constructing a motion vector between them. This forms a bitstream of run length encoded, quantized, discrete cosine transform macroblocks and of motion vectors. This bitstream is passed to and through an external buffer to a transmission medium. The number encoded bits read by a host from the external buffer is fed back to the encoder for calculation in real time of a dynamic buffer level indicator indicative of the fullness of the external buffer. The encoder may generate a BUFFEREMPTY flag, BUFFERALMOSTFULL flag and/or BUFFERFULL flag for the host.

On-chip dynamic buffer level indicators for digital video encoder

A control strategy is provided for dynamically encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel. The control strategy is a single pass strategy which allows individual encode bit rates to be dynamically adjusted for each video data stream based on part in relative complexity of the multiple streams of video data, as well as fullness of compressed video data buffers and a channel buffer coupled between the encoders and the constant bit rate channel. The control strategy includes encoding the multiple streams of video frames in parallel, and dynamically adapting encoding of at least one stream of the video frames based on relative complexity of the video frames. The bit rate for each stream of video frames is only changed at GOP boundaries, or if a scene change occurs. The calculated bit rate is preferably further modified based upon buffer fullness.

Control strategy for dynamically encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel

A method and system for encoding digital video picture data. In accordance with this method, the video picture data is partitioned into a group of blocks, at least some of those blocks are selected, one block at a time, and each of the selected blocks of data is encoded to form an encoded coefficient having an associated number of bits. The encoded coefficients are outputted, and an accumulated sum of the number of bits in the outputted encoded coefficients is kept. The outputting of the encoded coefficients is terminated at a defined time in order to prevent the accumulated sum from exceeding a given number.

Precise bit control apparatus with look-ahead for MPEG encoding

Simultaneous vertical spatial filtering and chrominance conversion is achieved with reduced data buffering and simplified filtering circuits by using a single filter stage and hybrid filter coefficients. Data latency is reduced and performance requirements are reduced while avoiding critical signal propagation paths. The filter and buffers are fully compatible with any scan format having consecutively presented lines of image data, including both progressive and interlaced scan formats.

Simultaneous vertical spatial filtering and chroma conversion in video images

Method, system and computer program product are provided for adaptively encoding in hardware, software or a combination thereof a series of still or partially still pictures using motion video encoding. A pre-encode statistics measurement unit is employed to derive statistics on each frame of the sequence of video frames to be encoded. The statistics are determinative of whether the frame comprises a still frame. If so, at least one controllable parameter to be used to encode the still frame is modified, and an encoding engine employs the at least one controllable parameter to encode the still frame so that pulsation artifacts are prevented between still frames of the series of still frames. Partial still picture encoding to prevent pulsation artifacts on a macroblock level is also addressed.

Adaptive encoding of a sequence of still frames or partially still frames within motion video

A control strategy is provided for dynamically encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel. The control strategy allows individual encode bit rates to be dynamically adjusted for each video data stream based in part on relative complexity of the multiple streams of video data, as well as fullness of compressed video data buffers and a channel buffer coupled between the encoders and the constant bit rate channel. The control strategy includes analyzing the multiple streams of video to determine relative complexity thereof, encoding the multiple streams of video frames in parallel, and dynamically adapting encoding of at least one stream of the video frames based on the relative complexity of the video frames. The bit rate for each stream of video frames is only changed at GOP boundaries, or if a scene change occurs. The calculated bit rate is preferably further modified based upon buffer fullness.

Adaptively encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel

A technique is provided for encoding macroblocks of a frame of a sequence of video frames initially employing luminance data only to analyze temporal redundancy of the macroblocks within the frame. Upon deciding to code at least one macroblock as a non-intra macroblock, the technique includes considering whether to switch the coding decision for the at least one macroblock from non-intra to intra by evaluating chrominance data of the at least one macroblock. The evaluating of the chrominance data can include determining whether chrominance difference data, obtained by comparing chrominance values of a current macroblock with a reference macroblock, is greater than a user set chrominance difference threshold, and if so then the technique includes switching the macroblock coding decision. As a further qualification, the switching might occur only if the chrominance difference data is also greater than the corresponding luminance difference data.

Growing community of inventors

Endwell, NY, United States of America

Agnes Yee Ngai