Nvidia Corporation

Picsurf, Inc.

Zoran Corporation

Teralogic, Inc.

Other

Vivante Corporation

Lefan Zhong

Charles K Chui

Ram Prabhakar

Hong-Ye Gao

Cahide Kiris

Rongxiang Yi

Neal Meininger

Ed Ahn

Harikrishna Madadi Reddy

Wei Sun

Ignatius B Tjandrasuwita

Leonardo Vainsencher

Visalakshi Vaduganathan

David Noel Babbage, Ii

Wei-Lun Kao

Halim Theny

James M Spring

Huiming Zhang

A dynamic AC prediction technique is implemented in a data partition mode which automatically disables AC prediction for encoding the current macroblock in the next packet when packet overflow occurs. Otherwise, when there is no overflow, AC prediction remains enabled to maintain compression efficiency. More particularly, in the preferred embodiment, a determination is first made whether a macroblock causes a packet overflow if it is encoded in the current packet. If so, a new packet is initiated into which the macroblock is encoded without AC prediction as the first macroblock. Otherwise, the macroblock with AC prediction remains in the current packet and a new macroblock is encoded.

Dynamic packet size control for MPEG-4 data partition mode

A more efficient motion estimation process that utilizes a plurality of predicted start points (e.g., two predicted start points) based on blocks adjacent to the current block together with other improvements and requires minimal system resources (e.g., hardware resources and CPU processing) in its hardware implementation is provided. More particularly, the motion estimation technique in accordance with the present invention performs a plurality of coarse searches (either sequentially or in parallel) using a plurality of predicted start positions followed by a fine search.

MPEG motion estimation based on dual start points

A VLC data transfer interface is presented that allows digital data to be packed and assembled according to a format selectable from a number of formats while the data is being transferred to a desired destination.

Variable-length coding data transfer interface

A more efficient and less memory-access intensive post-processing architecture to substantially reduce blocking and ringing artifacts from decompressed video image data is provided. The post-processing architecture in accordance with the present invention implements a 'pipelined' filtering process wherein a filter buffer, a blocking filter, and a ringing filter are serially connected to each other. The serial connection allows decompressed video image data to be first provided to a blocking filter for substantially reducing blocking artifacts and then the de-blocked video image data can be provided directly to a ringing filter for de-ringing without the need of an extra memory access.

MPEG artifacts post-processed filtering architecture

Methods and systems for compressing an image are described. A plurality of transformed and quantized values associated with a block of image data is accessed. The block corresponds to a position within the image. A count of the number of bits encoded during run-length encoding of the block is made. Run-length encoding of the block is concluded should the count reach a limit.

Methods and systems for rate control in image compression

An image process system tiles an image data array, processing the tiles in a predefined order. Each tile of image data is processed by applying a predefined family of transform layers to the tile of image data so as to generate successive sets of transform coefficients. Distinct groups of bit planes of the transform coefficients are stored in distinct bitstreams or distinct portions of a file.

Multi-resolution image data management system and method based on tiled wavelet-like transform and distinct bitstreams for distinct groups of bit planes

An image process system tiles an image data array, processing the tiles in a predefined order. Each tile of image data is processed by applying a predefined family of transform layers to the tile of image data so as to generate successive sets of transform coefficients. Each set of transform coefficients include edge coefficients positioned at outside boundaries of the set of transform coefficients and non-edge coefficients positioned at interior locations of the set of transform coefficients. The sets of transform coefficients include a last set of transform coefficients, produced by the last transform layer, and one or more earlier sets of transform coefficients. The transform filters used include a short edge transform filter that is applied to image data at boundaries of the tile and to coefficients positioned at and near boundaries of each of the earlier sets of transform coefficients so as to generate the edge coefficients, and a long interior filter that is applied to image data at interior locations of the tile and to coefficients at interior locations of the earlier sets of transform coefficients so as to generate the non-edge coefficients. The edge transform filter has a shorter filter support than the interior transform filter, and both the edge transform filter and the interior transform filter are applied only to image data within the tile and only to transform coefficients within the earlier sets of transform.

Multi-resolution image data management system and method based on tiled wavelet-like transform and sparse data coding

An image is divided into nonoverlapping tiles, and the tiles are processed in a predefined order. Each tile is processed by applying a predefined family of transform layers to the tile so as to generate successive sets of transform coefficients. The sets of transform coefficients correspond to spatial frequency subbands of the image. The subbands are grouped in accordance with the transform layer that generated them. For one or more respective groups of subbands one or more parameters are generated whose value is indicative of the density of image features in the tile. The tile is classified in accordance with the values of the one or more parameters. Based on the classification, a set of quantization factors for the tile are selected, and then the transform coefficients of the tile are scaled by the selected set of quantization factors to as to generate a set of quantized transform coefficients.

A system and method for encoding a two dimensional array of data utilizes a library having entries corresponding to a set of predefined two dimensional adaptive spline wavelet waveforms. Each predefined two dimensional adaptive spline wavelet waveform is formed by the superposition of one or more B-splines. The data encoding method identifies a set of best matches between the array of data and the predefined two dimensional adaptive spline wavelet waveforms by generating the inner product of the array of data and each of the predefined two dimensional adaptive spline wavelet waveforms. Each inner product is generated by FIR filtering the data with a corresponding set of FIR filter coefficients, and then determining which of the inner products have largest values. Once a set of best matches has been found, the data encoding method generates data representing the identified set of best matches. The generated data indicates for each match: one of the library entries, a position within the array of data at which the match was found, and a magnitude of the match. The data encoding method is computationally efficient because inner products are computed by FIR filtering.

System and method for encoding video data using computationally efficient adaptive spline wavelets

An image processing system tiles an image data array. Each tile of image data is processed by applying a family of transform layers to the tile of image data so as to generate successive sets of transform coefficients. Each set of transform coefficients include edge coefficients positioned at outside boundaries of the set of transform coefficients and non-edge coefficients positioned at interior locations of the set of transform coefficients. The transform filters include a short edge transform filter that is applied to image data at boundaries of the tile and to coefficients positioned at and near boundaries of each of the earlier sets of transform coefficients so as to generate the edge coefficients, and a long interior filter that is applied to image data at interior locations of the tile and to coefficients at interior locations of the earlier sets of transform coefficients so as to generate the non-edge coefficients.

Image compression based on tiled wavelet-like transform using edge and non-edge filters

A digital camera includes a data capture device and data processing circuitry for processing image data representing captured images. The data processing circuitry processes tiles of the image data in a predefined order to generated processed image data, which is then stored as a data image file. The tiles are nonoverlapping portions of the image data. Each tile of image data is processing by applying a predefined sequence of transform layers to the tile of image data so as to generate successive layers of transform coefficients. In a preferred embodiment, the transform layers are successive applications of a wavelet-like decomposition transform. While each tile is processed, a predefined set of edge transform coefficients from a plurality of the transform layers are saved in memory for use while processing neighboring tiles. Further, the step of processing each tile includes applying at least a plurality of the transform layers to both transform coefficients generated by a prior transform layer and corresponding ones of the edge transform coefficients that were previously saved in memory while processing tiles neighboring the tile being processed. However, for some tiles along the edge of the image, there will be no edge transform coefficients from previously processed tiles to be used while processing the current tile.

Memory saving wavelet-like image transform system and method for digital camera and other memory conservative applications

A unified system and method for encoding an array of data. If the data array is comprised of DCT data, then coefficients from corresponding positions in the data array are mapped into a common blocks in a second data array so as to group similarly valued coefficients. If the data array is comprised of wavelet data and the wavelet tile is greater than a predetermined size, then the wavelet tile coefficients are mapped into a second array so as to combine coefficients from the same wavelet family. After the DCT or wavelet coefficients have been mapped, the DC coefficients are encoded using a differential pulse code modulation (DPCM) process. The maximum number of bits required to represent any coefficient in each block family in the data array is determined. The difference between the maximum number of bits required to represent any coefficient in the entire data array and each of the block family maximums is determined and encoded. Each block family is then successively divided into an upper leftmost block and a sub-family, and difference values between maximum numbers of component block groupings are determined and encoded. This process is then repeated until the selected sub-family comprises a single block. Whenever a sub-family is processed, if the sub-family is entirely filled with zero data it is so identified in the output data and no further processing of the sub-family is required. After all of the sub-families have been processed, the blocks are bit mask coded. After all the coefficients in a block have been processed, the bit mask is encoded. When blocks of a predetermined size are encountered, the method outputs the value of the coefficients in the block. A data decoder and method retraces the encoded data so as to reverse the process performed by the encoding method.

System and method for unified DCT and wavelet data coding

A video data encoding system and method represents video frame image by an initial array of data. The image data array is processed by a wavelet transform function to produce an array of transform coefficients, sometimes called a processed image data array. The processed image data array is divided into subarrays and each subarray is encoded by generating a plurality of encoded representations of the subarray, and then selecting and outputting the smallest of the encoded subarray representations. The plurality of encoded representations are generated by encoding the subarray using a predefined coding technique; generating a differential array that is equal to the difference, on a coefficient by coefficient basis, between the subarray of transform coefficients and the corresponding subarray from a prior video frame; and encoding the differential array. Each encoded subarray is preceded by a flag to indicate if the encoded data represents the subarray or the differential array. Each subarray and differential subarray is encoded using a nested split coding technique.

Growing community of inventors

Santa Clara, CA, United States of America

Lefan Zhong