公开(公告)号：US08824812B2

公开(公告)日：2014-09-02

申请号：US13633182

申请日：2012-10-02

Applicant: Mediatek Inc.

Inventor： Han-Liang Chou ,  Tung-Hsing Wu ,  Kun-Bin Lee ,  Chi-Cheng Ju

IPC: G06K9/36 ,  H04N7/50

CPC classification number: H04N7/50 ,  H04N19/115 ,  H04N19/15 ,  H04N19/17 ,  H04N19/34 ,  H04N19/61

Abstract: A method and apparatus of image data compression and decompression are disclosed. According to an embodiment of the present invention, the compression method partitions the image data into access units and encodes each access unit into a bitstream according to a target bit budget. Each access unit is encoded using first data compression to generate a first bitstream and the residual data is further encoded using second data compression to generate a second bitstream if the first bitstream is smaller than the target bit budget. In one example, the second data compression comprises bit plane coding applied to bit plane-ordered data, wherein the bit plane-ordered data is generated by scanning from a most significant bit to a least significant bit of the residual data in a bit plane-wise order. The decompression method comprises steps to recover reconstructed data from the first and second bitstreams.

Abstract translation: 公开了一种图像数据压缩和解压缩的方法和装置。 根据本发明的实施例，压缩方法将图像数据划分为访问单元，并根据目标比特预算将每个访问单元编码为比特流。 使用第一数据压缩对每个访问单元进行编码以生成第一比特流，并且如果第一比特流小于目标比特预算，则使用第二数据压缩进一步编码残差数据以生成第二比特流。 在一个示例中，第二数据压缩包括应用于位平面有序数据的位平面编码，其中通过在位平面有序数据中从残留数据的最高有效位扫描到最低有效位来生成位平面有序数据， 明智的秩序。 解压缩方法包括从第一和第二比特流恢复重建数据的步骤。

公开(公告)号：US20140105514A1

公开(公告)日：2014-04-17

申请号：US14049229

申请日：2013-10-09

Applicant: MEDIATEK INC.

Inventor： Tung-Hsing Wu ,  Han-Liang Chou ,  Kun-Bin Lee ,  Chi-Cheng Ju

IPC: H04N7/26

CPC classification number: H04N19/194 ,  H04N19/124 ,  H04N19/127 ,  H04N19/146 ,  H04N19/176

Abstract: An image compression method has at least the following steps: receiving source pixel data of a plurality of blocks of a frame; when a lossless compression mode is enabled for the frame, bypassing a source quantization operation and applying a lossless compression kernel to source pixel data of each of the blocks; and when a lossy compression mode is enabled for the frame, applying the source quantization operation to the source pixel data of each of the blocks to generate input pixel data of each of the blocks, and applying the lossless compression kernel to the input pixel data of each of the blocks. For example, the source quantization operation employs an adaptive quantization parameter for each of the blocks such that a size of compressed data of the frame generated under the lossy compression mode does not exceed a bit budget.

Abstract translation: 图像压缩方法至少具有以下步骤：接收帧的多个块的源像素数据; 当对帧进行无损压缩模式时，绕过源量化操作并将无损压缩内核应用于每个块的源像素数据; 并且当为帧启用有损压缩模式时，将源量化操作应用于每个块的源像素数据，以生成每个块的输入像素数据，并将无损压缩内核应用于 每个块。 例如，源量化操作对于每个块采用自适应量化参数，使得在有损压缩模式下生成的帧的压缩数据的大小不超过位预算。

公开(公告)号：US11882277B2

公开(公告)日：2024-01-23

申请号：US17725574

申请日：2022-04-21

Applicant: MEDIATEK INC.

Inventor： Tung-Hsing Wu ,  Shih-Yu Chen ,  Jer-Ming Chang ,  Chih-Hao Chang ,  Han-Liang Chou

IPC: H04N19/117 ,  H04N19/176 ,  H04N19/182 ,  H04N19/80

CPC classification number: H04N19/117 ,  H04N19/176 ,  H04N19/182 ,  H04N19/80

Abstract: A video encoding method includes: during a first period, performing an encoding process upon a first block group of a current frame to generate a first block group bitstream; and during a second period, transmitting a second block group bitstream derived from encoding a second block group of the current frame, wherein the second period overlaps the first period. The encoding process includes: during a first time segment of the first period, performing a first in-loop filtering process upon a first group of pixels; and during a second time segment of the first period, performing a second in-loop filtering process upon a second group of pixels, wherein the second time segment overlaps the first time segment, and a non-zero pixel distance exists between a first edge pixel of the first group of pixels and a second edge pixel of the second group of pixels in a filter direction.

公开(公告)号：US10511859B2

公开(公告)日：2019-12-17

申请号：US15404313

申请日：2017-01-12

Applicant: MEDIATEK INC.

Inventor： Li-Heng Chen ,  Tung-Hsing Wu ,  Han-Liang Chou

IPC: H04N19/593 ,  H04N19/174 ,  H04N19/182 ,  H04N19/105 ,  H04N19/147 ,  H04N19/91 ,  H04N19/124 ,  H04N19/14

Abstract: A method and apparatus of image coding including a no-residue prediction mode are disclosed. At the encoder side, current predictors for a current coding unit are determined according to a target prediction process associated with a target prediction mode. The current predictors are determined based on pixel data including reconstructed pixels in the current slice or the current image. The target prediction process is applied to the current coding unit using the current predictors to generate prediction residues. A distortion condition or a cost measure for the prediction residues is determined. If the distortion condition is satisfied or the cost measure favors coding without the prediction residues, the current coding unit is encoded into current compressed bits of the current coding unit without compressed data associated with the prediction residues. A corresponding decoding method and apparatus are also disclosed.

公开(公告)号：US20190281320A1

公开(公告)日：2019-09-12

申请号：US16280039

申请日：2019-02-20

Applicant: MEDIATEK INC.

Inventor： Tung-Hsing Wu ,  Ting-An Lin ,  I-Hsuan Lee ,  Han-Liang Chou

IPC: H04N19/60 ,  H04N19/88 ,  H04N19/119 ,  H04N19/184 ,  H04N19/176

Abstract: An encoding method is used for encoding an image. The image includes a plurality of blocks each having a plurality of pixels. The encoding method includes: encoding a plurality of data partitions of block data of a block in the image to generate a plurality of compressed bitstream segments, respectively; and combining the compressed bitstream segments to generate an output bitstream of the block. A bit group based interleaving process is involved in generating the output bitstream. According to the bit group based interleaving process, each of the compressed bitstream segments is divided into a plurality of bit groups each having at least one bit, and the output bitstream includes consecutive bit groups belonging to different compressed bitstream segments, respectively.

公开(公告)号：US20190007680A1

公开(公告)日：2019-01-03

申请号：US16004419

申请日：2018-06-10

Applicant: MEDIATEK INC.

Inventor： Li-Heng Chen ,  Lien-Fei Chen ,  Tung-Hsing Wu ,  Han-Liang Chou

IPC: H04N19/119 ,  H04N19/174 ,  H04N19/159 ,  H04N19/184 ,  H04N19/147 ,  H04N19/91 ,  H04N19/436

Abstract: A video encoder includes a video encoding circuit and a slice decision circuit. The video encoding circuit encodes a first slice in a frame according to a first coding unit boundary between an end of the first slice and a start of a second slice in the frame, and outputs a first bitstream of the first slice. The slice decision circuit predicts the first coding unit boundary before a bitstream of a last coding unit of the first slice is generated by the video encoding circuit, and informs the video encoding circuit of the first coding unit boundary. The video encoding circuit refers to the first coding unit boundary predicted by the slice decision circuit to ensure that a bitstream size of the first bitstream is constrained by a predetermined bitstream size threshold.

公开(公告)号：US10142640B2

公开(公告)日：2018-11-27

申请号：US14531930

申请日：2014-11-03

Applicant: MEDIATEK INC.

Inventor： Tung-Hsing Wu ,  Kun-Bin Lee

IPC: H04N19/43 ,  H04N19/52 ,  H04N19/56 ,  H04N19/44

Abstract: A buffer device is used by a compressor/decompressor with block prediction for storing information needed to encode/decode a pixel line. Vector information (e.g., final block prediction vector information and/or partial block prediction vector information) of at least one later pixel group is computed while a current pixel group is being encoded/decoded. The buffer device does not need to store vector information of all pixel groups of one pixel line at the same time. For example, one pixel line is composed of M pixel groups; and when the compressor/decompressor is encoding/decoding the current pixel line, final/partial vector information sets of at most K pixel groups are allowed to be stored in the buffer device simultaneously, where K is a positive integer smaller than M.

公开(公告)号：US10038904B2

公开(公告)日：2018-07-31

申请号：US15031281

申请日：2014-10-24

Applicant: MEDIATEK INC.

Inventor： Kun-Bin Lee ,  Tung-Hsing Wu ,  Han-Liang Chou

IPC: H04N19/119 ,  H04N19/436 ,  H04N19/184 ,  H04N19/44 ,  H04N19/463 ,  H04N19/172 ,  H04N19/174 ,  H04N21/43 ,  H04N19/115 ,  H04N21/41 ,  H04N5/08

CPC classification number: H04N19/119 ,  H04N5/08 ,  H04N19/115 ,  H04N19/172 ,  H04N19/174 ,  H04N19/184 ,  H04N19/436 ,  H04N19/44 ,  H04N19/46 ,  H04N19/463 ,  H04N21/4122 ,  H04N21/4302

Abstract: An image processing method includes at least the following steps: partitioning a picture into a plurality of slices, wherein each slice row in the picture includes at least one slice; generating a compressed picture by encoding each of the slices; and controlling at least one of start of transmission of encoded data of a slice row and end of the transmission of encoded data of the slice row according to a transmission synchronization event.

公开(公告)号：US20170339423A1

公开(公告)日：2017-11-23

申请号：US15595933

申请日：2017-05-15

Applicant: MEDIATEK INC.

Inventor： Li-Heng Chen ,  Tung-Hsing Wu ,  Han-Liang Chou

IPC: H04N19/50 ,  H04N19/70 ,  H04N19/436 ,  H04N19/124 ,  H04N19/103 ,  H04N19/184 ,  H04N19/176 ,  H04N19/182 ,  H04N19/186

CPC classification number: H04N19/176 ,  H04N19/103 ,  H04N19/147 ,  H04N19/436

Abstract: One exemplary image encoding method for encoding an image includes following steps: calculating a mean value of each color channel of a plurality of reconstructed pixels; determining a first predictor used by a first candidate coding mode of a current coding block according to mean values of color channels of the reconstructed pixels; determining a second predictor used by a second candidate coding mode of the current coding block according to the mean values of the color channels of the reconstructed pixels, wherein determining the first predictor and determining the second predictor are performed in a parallel manner; determining a coding mode selected from candidate coding modes including at least the first candidate coding mode and the second candidate coding mode; and encoding the current coding block into a part of a bitstream according to at least the determined coding mode.

公开(公告)号：US20170311003A1

公开(公告)日：2017-10-26

申请号：US15490920

申请日：2017-04-19

Applicant: MEDIATEK INC.

Inventor： Tung-Hsing Wu ,  Li-Heng Chen ,  Han-Liang Chou

IPC: H04N19/91 ,  H04N19/124 ,  H04N19/182 ,  H04N19/44

CPC classification number: H04N19/91 ,  H04N19/124 ,  H04N19/182 ,  H04N19/436 ,  H04N19/44

Abstract: An entropy encoding method includes following steps: receiving symbols of a pixel group; entropy encoding data derived from the symbols of the pixel group to generate a first bitstream portion and a second bitstream portion, wherein the first bitstream portion includes encoded magnitude data of the symbols of the pixel group, and the second bitstream portion includes encoded sign data of at least a portion of the symbols of the pixel group; and generating a bitstream segment of the pixel group by combining at least the first bitstream portion and the second bitstream portion. The pixel group includes a plurality of pixels, and the symbols of the pixel group have a plurality of symbol values of the pixels, respectively. When a first symbol value has a zero magnitude value, a sign value of the first symbol value is not entropy encoded into the second bitstream portion.