Abstract: System or methods may include loading information associated with a window of pixels from memory into a matrix of processing elements, the window of pixels being selected from a plurality of pixels associated with a reference frame, determining a current frame comprising a plurality of pixels, determining a first block of pixels from the plurality of pixels of the current frame, comparing each pixel of the first block of pixels to each pixel of the window of pixels by comparing the first block of pixels to portions of the window of pixels in an exhaustive manner, wherein comparing comprises calculating a first deviation result for each pair of compared pixels and aggregating the first deviation results for each position of the first block of pixels in the window of pixels, the calculating being performed by the matrix of processors.

Abstract: A method of decoding a stream of compression-encoded image data. The method includes supplying at least two values to an adder. At least one of the values is determined based at least in part on a type of a current syntax element in the stream of compression-encoded image data. The method also includes adding the at least two values at the adder to produce an index value. The method further includes using the index value to access a look-up table.

Abstract: An apparatus includes a compact table generator module and a storage medium. The compact table generator creates a compact code table having multiple code word groups arranged in one or more hierarchies. Each of these code word groups includes a group mask that corresponds to one or more code words in a variable length code table. The storage medium stores information regarding the code word groups, such as a group mask for each of the code word groups, and a code index for each code word in the variable length code table.

Abstract: In some embodiments, a method includes generating a first image data frame, a first error image data frame and a second error image data frame based at least in part on a bitstream. The method further includes upscaling or de-quantizing the first image data frame to provide a first predicted image data frame, generating a second image data frame based at least in part on the first predicted image data frame and the first error image data frame, upscaling or de-quantizing the second image data frame to provide a second predicted image data frame, and generating a third image data frame based at least in part on the second predicted image data frame and the second error image data frame. In some embodiments, an apparatus includes a storage medium having stored instructions that when executed by a machine result in the method.

Abstract: An embodiment improves the operation of a H.264 and Joint Scalable Video Codec (e.g., JSVC/H.264 Amendment 3) video decoder by managing neighboring block data during the decoding process. An embodiment pre-computes neighboring block tables to efficiently locate the neighboring block data required to decode a current macroblock. In particular, the pre-computed most probable joint neighboring block tables disclosed herein handle both macroblock adaptive frame field (MBAFF) coding and non-MBAFF coding. An embodiment is further capable of managing variable block sizes. Other embodiments are described and claimed.

Abstract: A binary arithmetic decoding apparatus includes first, second and third pairs of look-up tables and first, second and third multiplexers. The first multiplexer selects between the respective outputs of the two look-up tables of the first pair of look-up tables. The second multiplexer selects between the respective outputs of a first look-up table of the second pair of look-up tables and of a first look-up table of the third pair of look-up tables. The third multiplexer selects between the respective outputs of a second look-up table of the second pair of look-up tables and of a second look-up table of the third pair of look-up tables. The three multiplexers are controlled in common.

Abstract: System or methods may include loading information associated with a window of pixels from memory into a matrix of processing elements, the window of pixels being selected from a plurality of pixels associated with a reference frame, determining a current frame comprising a plurality of pixels, determining a first block of pixels from the plurality of pixels of the current frame, comparing each pixel of the first block of pixels to each pixel of the window of pixels by comparing the first block of pixels to portions of the window of pixels in an exhaustive manner, wherein comparing comprises calculating a first deviation result for each pair of compared pixels and aggregating the first deviation results for each position of the first block of pixels in the window of pixels, the calculating being performed by the matrix of processors.

Abstract: A method of decoding a stream of compression-encoded image data. The method includes supplying at least two values to an adder. At least one of the values is determined based at least in part on a type of a current syntax element in the stream of compression-encoded image data. The method also includes adding the at least two values at the adder to produce an index value. The method further includes using the index value to access a look-up table.

Abstract: A method may include performing an N bit-at-a-time matching operation for a first N bits in an encoded input stream of bits using a lookup table. The matching operation may obtain a first address in the table, and N may be an integer greater than one. The method may also include obtaining a second address in the table based on a mask and a jump address that are associated with the first address and a second number of bits in the encoded input stream. An index value may be output based on the second address in the table.

Abstract: A method may include performing an N bit-at-a-time matching operation for a first N bits in an encoded input stream of bits using a lookup table. The matching operation may obtain a first address in the table, and N may be an integer greater than one. The method may also include obtaining a second address in the table based on a mask and a jump address that are associated with the first address and a second number of bits in the encoded input stream. An index value may be output based on the second address in the table.

Abstract: According to some embodiments, a first value of a first parameter type is received in connection with a macroblock that represents a portion of an image. The macroblock may be divided into a first set of sub-portions, and different values of the first parameter type may be associated with different sub-portions of the first set. The first value may then be stored in a context buffer that includes a first context area associated with the first parameter type and a second context area associated with a second parameter type. The first context area might be, for example, adapted to store fewer values for each parameter type as compared to the second context area.

Abstract: According to some embodiments, an external memory unit stores rows of macroblocks representing an image. A set of macroblocks may be transferred from the external memory unit to a local buffer, the set of macroblocks including fewer macroblocks than a row. A macroblock may then be decoded based on information in the local buffer to generate an image signal.