摘要:
A video encoding/decoding apparatus and method are disclosed. The video encoding/decoding apparatus includes: a video encoder for generating predicted subblocks by predicting subblocks into which macroblock is divided, generating residual subblocks by subtracting predicted subblocks from subblocks, generating quantized transform coefficients by transforming and quantizing residual subblocks by using transform selected according to subblock size, and generating encoded video data by using different context tables according to the macroblock and subblocks when using CABAC to encode the quantized transform coefficients; and a video decoder for reconstructing information of partition type, motion, transform type and quantized transform coefficients by decoding encoded data by using a CABAC and different context tables according to a macroblock and subblocks, reconstructing residual subblocks by inverse-quantizing/inverse-transforming the quantized transform coefficients by using a transform identified by the transform type information, generating predicted subblocks, reconstructing the subblocks, and reconstructing the macroblock based on the subblocks.
摘要:
A video encoding/decoding apparatus including a video encoder for generating a prediction block, generating a residual block by subtracting the prediction block from the current block. The video encoder is for generating a frequency-transformed block by transforming and quantizing the residual block, generating a masked frequency-transformed block by masking the frequency-transformed block by using one or more frequency mask tables. The video encoder is for encoding information about a frequency mask table used for masking the masked frequency-transformed block. The apparatus includes a video decoder for extracting a masked quantized frequency coefficient string recorded in the bitstream. The video decoder is for generating a quantized frequency coefficient string by zero setting all quantized frequency coefficients at positions of no records in the bitstream from a masked quantized frequency coefficient string. The video decoder is for generating a quantized frequency-transformed block by inversely scanning the generated quantized frequency coefficient string.
摘要:
Disclosed is a high-resolution video encoding/decoding method and apparatus. The video encoding method includes: predicting a current block to generate a predicted block, subtracting the predicted block from the current block to generate a residual block, determining a transform and quantization type according to the block type of the current block; transforming and quantizing the residual block according to the determined transform and quantization type; and encoding the transformed and quantized residual block. According to the video encoding/decoding method and apparatus, not only the encoding efficiency can be improved because it enables an encoding using a high correlation between temporally/spatially adjacent pixels appearing in a video, but the compression efficiency can also be improved by reducing block distortion.
摘要:
The present disclosure relates to a video encoding/decoding apparatus and method, in which skip information indicating whether a block is a skip block is encoded, partition information of the block and skip motion information of the block are encoded or prediction information of the block containing the partition information of the block and intra prediction mode information or motion information are encoded according to the skip information, residual signal information of the block is predictive-encoded based on the prediction information and the transform information, and an encoded signal is reconstructed. The method and the apparatus can improve the video compression efficiency by efficiently encoding the encoding information used for the video encoding and selectively using various encoding methods and decoding methods in encoding the video.
摘要:
Image encoding/decoding apparatus and method are disclosed. The image encoding apparatus includes: a target block splitter for splitting a target block of an input image to be encoded into a plurality of subblocks according to positions of pixels; a key block encoding unit for selecting a random key block, the size of which is larger than a pixel size of at least one of the plurality of subblocks split by the target block splitter, and performing an intra-prediction or inter-prediction encoding process on the selected key block; and a non-key block encoding unit for designating the rest of the subblocks, except for the selected key block, as a non-key block, and performing a spatial prediction process on the designated non-key block by using a key block reconstructed by the key block encoding unit and values of peripheral pixels of a current block.
摘要:
A video encoding/decoding apparatus and method are disclosed. The video encoding/decoding apparatus includes: a video encoder for encoding a differential value between a quantization parameter value of a macroblock of a video and a quantization parameter value of a subblock in an extended macroblock based on additional information on the macroblock and the subblock; and a video decoder for reconstructing a differential value of a quantization parameter encoded by the video encoder, reconstructing a residual block based on a reconstructed differential value of the quantization parameter, and reconstructing the video based on the residual block reconstructed.
摘要:
A video encoding apparatus includes a candidate block type configuration unit for configuring block type candidates for encoding an image by using variable sized macroblocks; a video encoder for encoding an input image for each block type candidate; and a block type determiner for determining a block type based on encoding costs by the block type candidate, and generating a bitstream including image data encoded by the determined block type and information for the determined block type. Determining the block type to match the image characteristics enables combinations of diverse modes such as a large block and macroblock size, a minimum subblock size, available divisional block type, prediction mode type within a macroblock, and so on to achieve more effective encoding and decoding of high resolution images.
摘要:
Disclosed is a high-resolution video encoding/decoding method and apparatus. The video encoding method includes: predicting a current block to generate a predicted block, subtracting the predicted block from the current block to generate a residual block, determining a transform and quantization type according to the block type of the current block; transforming and quantizing the residual block according to the determined transform and quantization type; and encoding the transformed and quantized residual block. According to the video encoding/decoding method and apparatus, not only the encoding efficiency can be improved because it enables an encoding using a high correlation between temporally/spatially adjacent pixels appearing in a video, but the compression efficiency can also be improved by reducing block distortion.
摘要:
Video encoding/decoding apparatus includes video encoder and video decoder. Video encoder generates predicted subblocks by predicting subblocks into which macroblock is divided, generates residual subblocks by subtracting predicted subblocks from subblocks, generates quantized transform coefficients by transforming and quantizing residual subblocks by using transform selected according to size of subblock, generates encoded video data by encoding quantized transform coefficients, and selectively calculates and determines encoding cost when determining encoding mode for at least one of motion vector, transform type, prediction mode, and block type for subblocks. Video decoder reconstructs partition type information, motion information, transform type information, and quantized transform coefficients by decoding encoded data, reconstructs residual subblocks by inverse-quantizing and inverse-transforming quantized transform coefficients by using transform identified by transform type information, generates predicted subblocks by predicting subblocks identified according to partition type information by using motion information, and reconstructs subblocks by adding predicted subblocks and residual subblocks.
摘要:
The present disclosure relates to a method and apparatus for encoding/decoding a motion vector and a method and apparatus for encoding/decoding video using same. The motion vector encoding method includes selecting a predicted motion vector candidate set including one or more predicted motion vector candidates for a block; determining one or more search ranges for predicted motion vector candidate set; selecting one predicted motion vector candidate among one or more predicted motion vector candidates as predicted motion vector for each search point with respect to each search point within search range by first determination criterion prearranged with video decoding apparatus; selecting one predicted motion vector among the predicted motion vectors for each search point by a second determination criterion not prearranged with the video decoding apparatus, and determining predicted motion vector, differential motion vector, and current motion vector; and generating and encoding the differential motion vector as motion information.