摘要:
A block transform-based digital media codec achieves faster performance by re-mapping components of the digital media data into vectors or parallel units on which many operations of the transforms can be performed on a parallel or single-instruction, multiple data (SIMD) basis. In the case of a one-dimensional lapped biorthogonal transform, the digital media data components are re-mapped into vectors on which butterfly stages of both overlap pre-/post-filter and block transform portions of the lapped transform can be performed on a SIMD basis. In the case of a two-dimensional lapped biorthogonal transform, the digital media data components are re-mapped into vectors on which a Hadamard operator of both overlap pre-/post-filter and block transform can be performed on a SIMD basis.
摘要:
A system, a method and computer-readable media for organizing a bitstream of compressed data that represents an image. The image may be partitioned into independently decodable regions. The portion of the compressed bitstream associated a selected region is decoded. This decoding yields a series of transform coefficients. Areas of the image that surround the selected region are identified, and information associated with these areas is decoded to yield additional transform coefficients. The original series of transform coefficients and the additional transform coefficients are used to reconstruct the selected region of the image.
摘要:
A block transform-based digital media codec uses a transform coefficient prediction that takes into account a dominant directionality of the digital media data (e.g., an image with strong horizontal or vertical features), and further operates compatibly with a two-stage transform. For DC and DCAC coefficients from an inner stage transform of a macroblock, the codec calculates and compares directionality metrics based on inner stage transform DC coefficients of neighboring macroblocks to determine dominant directionality. For DCAC coefficients from an outer stage transform of blocks within the macroblock, the codec calculates and compares directionality metrics based on the inner stage transform DCAC coefficients of the macroblock to detect dominant directionality. The determination of directional dominance can also take into account information from other channels (e.g., chrominance as well as luminance).
摘要:
A digital media encoder/decoder uses a flexible quantization technique that provides the ability to vary quantization along various dimensions of the encoded digital media data, including spatial, frequency sub bands and color channels. The codec utilizes a signaling scheme to signal various permutations of flexible quantization combinations efficiently for primary usage scenarios. When a choice of quantizer is available, the codec efficiently encodes the current quantizer by defining a subset of quantizers and indexes the current quantizer from the set.
摘要:
An efficient lapped transform is realized using pre- and post-filters (or reversible overlap operators) that are structured of unit determinant component matrices. The pre- and post-filters are realized as a succession of planar rotational transforms and unit determinant planar scaling transforms. The planar scaling transforms can be implemented using planar shears or lifting steps. Further, the planar rotations and planar shears have an implementation as reversible/lossless operations, giving as a result, a reversible overlap operator.
摘要:
An efficient lapped transform useable in digital media codecs is realized using a reversible 2-dimensional overlap operator for pre/post filtering that is applied on a staggered grid relative to a core transform. The 2-dimensional lapped operator is based on a separably applied 1-dimensional reversible lapped operator, which is re-arranged as a sequence of elementary transforms in interleaved stages on subsets of points of the respective block and further implemented in lifting steps for computational efficiency. The interleaved stages include applying rotation and scaling stages between initial and final stages involving a normalized 2×2 Hadamard transform.
摘要:
A block transform-based digital media codec achieves faster performance by re-mapping components of the digital media data into vectors or parallel units on which many operations of the transforms can be performed on a parallel or single-instruction, multiple data (SIMD) basis. In the case of a one-dimensional lapped biorthogonal transform, the digital media data components are re-mapped into vectors on which butterfly stages of both overlap pre-/post-filter and block transform portions of the lapped transform can be performed on a SIMD basis. In the case of a two-dimensional lapped biorthogonal transform, the digital media data components are re-mapped into vectors on which a Hadamard operator of both overlap pre-/post-filter and block transform can be performed on a SIMD basis.
摘要:
An efficient lapped transform useable in digital media codecs is realized using a reversible 2-dimensional overlap operator for pre/post filtering that is applied on a staggered grid relative to a core transform. The 2-dimensional lapped operator is based on a separably applied 1-dimensional reversible lapped operator, which is re-arranged as a sequence of elementary transforms in interleaved stages on subsets of points of the respective block and further implemented in lifting steps for computational efficiency. The interleaved stages include applying rotation and scaling stages between initial and final stages involving a normalized 2×2 Hadamard transform
摘要:
An efficient lapped transform is realized using pre- and post-filters (or reversible overlap operators) that are structured of unit determinant component matrices. The pre- and post-filters are realized as a succession of planar rotational transforms and unit determinant planar scaling transforms. The planar scaling transforms can be implemented using planar shears or lifting steps. Further, the planar rotations and planar shears have an implementation as reversible/lossless operations, giving as a result, a reversible overlap operator.
摘要:
A system, a method and computer-readable media for organizing a bitstream of compressed data that represents an image. The image may be partitioned into independently decodable regions. The portion of the compressed bitstream associated a selected region is decoded. This decoding yields a series of transform coefficients. Areas of the image that surround the selected region are identified, and information associated with these areas is decoded to yield additional transform coefficients. The original series of transform coefficients and the additional transform coefficients are used to reconstruct the selected region of the image.