摘要:
A motion refinement engine can be used in a video encoder for encoding a video input signal that includes a sequence of images that are segmented into a plurality of macroblocks. The motion refinement engine includes a motion search module, that generates at least one motion search motion vector for a macroblock of the plurality of macroblocks based on a first plurality of directions. A motion refinement module generates at least one refined motion vector for the macroblock of the plurality of macroblocks, based on a second plurality of directions, wherein the second plurality of directions differ from the first plurality of directions. In a further embodiment, motion search, motion refinement, and motion compensation are selectively performed in a single pass.
摘要:
A motion compensation module can be used in a video encoder for encoding a video input signal that includes a sequence of images that are segmented into a plurality of macroblocks. The motion compensation module includes a shared memory and a motion search module that generates a motion search motion vector for a plurality of subblocks for a plurality of partitionings of a macroblock of a plurality of macroblocks and generates a selected group of the plurality of partitionings and stores the selected group of partitionings and the corresponding motion search motion vectors in the shared memory. A motion refinement module retrieves the selected group of partitionings and the corresponding motion search motion vectors from the shared memory, that generates a refined motion vector for the plurality of subblocks for the selected group of the plurality of partitionings of the macroblock of the plurality of macroblocks, based on the motion search motion vector for each of the plurality of subblocks of the macroblock of the plurality of macroblocks.
摘要:
A motion compensation module can be used in a video encoder for encoding a video input signal that includes a sequence of images that are segmented into a plurality of macroblocks. The motion compensation module includes a shared memory and a motion search module that generates a motion search motion vector for a plurality of subblocks for a plurality of partitionings of a macroblock of a plurality of macroblocks and generates a selected group of the plurality of partitionings and stores the selected group of partitionings and the corresponding motion search motion vectors in the shared memory. A motion refinement module retrieves the selected group of partitionings and the corresponding motion search motion vectors from the shared memory, that generates a refined motion vector for the plurality of subblocks for the selected group of the plurality of partitionings of the macroblock of the plurality of macroblocks, based on the motion search motion vector for each of the plurality of subblocks of the macroblock of the plurality of macroblocks.
摘要:
A video encoder includes a forward path section that processes a video input signal in accordance with a video compression standard, based on a plurality of reference pictures, to generate a processed video signal and a plurality of transformed quantized residual pixel values. A decoding engine decodes the plurality of transformed quantized residual pixel values in accordance with the video compression standard to the plurality of reference pictures.
摘要:
A multi-format video decoder includes an entropy decoding device that generates entropy decoded (EDC) data from an encoded video signal. A multi-format video decoding device includes a memory module that stores a plurality of operational instructions including at least one vector read instruction that includes vector read orientation data. A plurality of vector processor units generate a decoded video signal from the EDC data, wherein at least one of the plurality of vector processors include a matrix memory that stores matrix data corresponding to a plurality of rows and columns and that generates vector read data in a first read orientation when the vector read orientation data has a first value and that generates vector read data in a second read orientation when the vector read orientation data has a second value.
摘要:
A video decoder includes an entropy decoding device that generates entropy decoded (EDC) data from an encoded video signal. A multi-format video decoding device includes a plurality of vector processor units that generate a decoded video signal from the EDC data. The plurality of vector processing units are programmed via VPU instructions formatted to include a vector instruction portion, a scalar instruction portion, and a branching instruction portion.
摘要:
A multi-format video decoder includes an entropy decoding device that generates entropy decoded (EDC) data from an encoded video signal. A multi-format video decoding device includes a memory module that stores format configuration data corresponding to a plurality of video coding formats. A plurality of vector processor units generate a decoded video signal from the EDC data, wherein the plurality of vector processing units are configured, based on the configuration data, to a selected one of the plurality of video coding formats.
摘要:
A motion compensation module can be used in a video encoder for encoding a video input signal that includes a sequence of images that are segmented into a plurality of macroblocks. The motion compensation module includes a motion search module, that generates a motion search motion vector for a plurality of subblocks for a plurality of partitionings of a macroblock of a plurality of macroblocks based on a sum of accumulated differences (SAD) cost. A motion refinement module, when enabled, generates a refined motion vector for the plurality of subblocks for the plurality of partitionings of the macroblock of the plurality of macroblocks, based on the motion search motion vector for each of the plurality of subblocks of the macroblock of the plurality of macroblocks and based on a sum of accumulated transform differences (SATD) cost. A mode decision module operates on either SAD costs or SATD costs, based on whether the refinement module is enabled or disabled.
摘要:
A video decoder includes an entropy decoding device that generates entropy decoded (EDC) data from an encoded video signal. A multi-format video decoding device includes a plurality of vector processor units that generate a decoded video signal from the EDC data. The plurality of vector processing units includes at least one filter vector processor that operates in conjunction with a plurality of programmable filter parameters.
摘要:
A multi-format video decoder includes an entropy decoding device that generates entropy decoded (EDC) data from an encoded video signal. A multi-format video decoding device includes a memory module that stores format configuration data corresponding to a plurality of video coding formats. A plurality of vector processor units generate a decoded video signal from the EDC data, wherein at least one of the plurality of vector processors include a vector function module that generates vector function data based on a vector function of a first input vector and a second input vector. A selection module selects each element of a vector output as one of: a corresponding element of the vector function data, and a corresponding element of a third input vector.