International Business Machines Corporation

Charles John Stein

Charles E Boice

Edward Francis Westermann

Agnes Yee Ngai

Everett G Vail, Iii

Krishna Chaitanya Ratakonda

Barbara A Hall

Lawrence D Curley

Ronald S Svec

Chuck Hong Ngai

Adrian Stephen Butter

John Mark Kaczmarczyk

Jeffery Dean Carr

Stephen Philip Pokrinchak

A method and apparatus are provided for implementing B-picture scene changes. A prediction stage predicts a B-picture scene change based upon a sequence of statistical information in an encoder order and a reaction stage is responsive to the prediction stage for modifying a quantization scale of a rate control algorithm.

Implementing B-picture scene changes

Method and apparatus for implementing B-picture scene changes

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

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

Method, system and computer program product are provided for real time evaluation of compressed picture quality, in hardware, software or a combination thereof, during encoding of a sequence of video data. A received video signal is encoded by quantizing the signal after frequency transformation to produce a quantized video signal which is automatically evaluated in real time for data loss. Loss statistics, representative of picture quality of the quantized video signal, are generated for comparison with, for example, a predefined threshold as a dynamic control over outputting of the quantized video signal to the encode pipeline. As an enhancement, multiple quantizers are employed in parallel, each using a different quantization scale to produce multiple different quantization signals. The multiple quantization signals are evaluated in real time for picture quality and an optimum signal is selected for forwarding within the encode pipeline.

Real-time evaluation of compressed picture quality within a digital video encoder

A digital video encoder system having a motion estimation processor, and an interface to the motion estimation processor. The motion estimation processor includes a reference memory interface, and inverse quantization processor, an inverse discrete cosine transform processor, and a motion estimation processor unit including a hierarchal search unit. The motion estimation processor is utilized generating temporally compressed datastreams, that is, I-P and I-P-B datastreams.

Motion estimation processor for a digital video encoder

An MPEG digital video decoder system, method and computer program product are presented for monitoring decoding of an encoded digital video signal for one or more predefined illegal conditions. Error detection logic is coupled to the variable length (VLC) decoder, inverse quantizer (IQ), inverse discrete cosine transformer (IDCT) and motion compensator (MC) of the decoder for detecting an illegal condition within at least one of the VLC decoder, IQ, IDCT and MC during decoding of the encoded digital video signal. The monitored illegal conditions can include a VLC/IQ control error, an IQ level overrun, and IQ/IDCT buffer error, an MC idle error and an MC macroblock start error. Error signals are reported to a central error register which is monitored periodically by the decoder's control processor. The control processor initiates recovery within the decoder system prior to stoppage of the system due to the illegal condition.

Early error detection within an MPEG decoder

A method of and apparatus for identifying a Dual Prime motion estimation best match and generating motion vectors pertaining thereto for inter-picture video compression in a motion picture having images of F.sub.1 and F.sub.2 parities. The Dual Prime method of motion estimation described herein includes a method of generating motion vectors. The motion vectors point from a macroblock in a current field to a macroblock in a past field for inter-picture video compression in a motion picture having images of F.sub.1 and F.sub.2 parities. The first step in the method is defining a macroblock in a parity field of the current picture. Next, the preceding field of the same or opposite parity is searched to find a first best match macroblock in the preceding field. Once a best match macroblock is found, a vector is formed from the current macroblock in the current parity field to the first best match macroblock in the preceding parity field. This vector is scaled so that it refers to a second dual prime macroblock in the opposite parity field, that is the parity field of opposite parity with respect to the first best match macroblock which defines the center of a dual prime search window. Then, the average of the first best match macroblock and the set of second dual prime macroblocks in the dual prime search window is taken and the search window is searched to determine the best match dual prime macroblock. The search is typically conducted at half pixel locations inside a one square pixel dual prime search window. The appropriate dual prime motion vector and differential motion vector (dmv) are formed. These are motion vectors which point to both the first best match macroblock and the best match dual prime macroblock.

Dual prime motion estimation system and method

A digital video encoding method, apparatus and computer program product are described with enhanced encoding of a picture having a plurality of macroblocks. The approach is to partition the picture into at least two regions and then to set or adjust at least one encoding parameter for each macroblock of the picture based upon the region of the picture within which the macroblock is located. As one specific implementation, the picture is partitioned into a center region and at least one outer region and the at least one encoding parameter is set to enhance picture quality in the center region at the expense of picture quality in the at least one outer region. The encoding parameter can, for example, be a motion estimation parameter, a quantization value, or a target bits per macroblock.

Growing community of inventors

Peckville, PA, United States of America

Charles John Stein