Abstract: Methods for processing of video sequences that may contain telecined (3:2 pull down) frame sequences are provided. A method for detecting 3:2 pull down is provided that measures vertical detail in frames of a video sequence and uses the variation in vertical detail over time to decide whether the video sequence contains normal interlace content or 3:2 pull down content. A method for improving the compression of detected 3:2 pull down content is also provided that controls the selection of field or frame coding mode for frames of 3:2 pull down content and the selection of reference fields for encoding duplicated fields in the 3:2 pull down content.

Abstract: A method of video coding comprising Intra prediction coding at least a predetermined constant M macroblocks within an ordinary frame. These M macroblocks include naturally selected macroblocks where the bit cost of Intra prediction coding is less than for Inter prediction coding and optional additional macroblocks. The optional additional macroblocks have the least bit cost of Inter prediction coding as adjusted by a penalty factor each time a corresponding macroblock is Intra coded in a predetermined number of frames. An end frame Intra prediction codes each macroblock having no corresponding Intra prediction coded macroblock in a predetermined number of ordinary frames.

Abstract: A method for conserving memory accesses in motion compensation of compressed video gathers a plurality of motion compensation data accesses into a single bounding box for direct memory access data transfer. The method places the motion compensation data of a current macroblock into a current bounding box until placing the motion compensation data into the current bounding box causes the ratio of motion compensation data to bounding box data to fall below a predetermined memory utilization threshold.

Abstract: A video encoding method determines the best video encoding mode for a macroblock in the SKIP mode and comparing this cost with other modes. This avoids sub-pixel interpolations for fractional pixels. This models the cost function in a quadratic model and computes the cost for a nearest integer pel and plural adjacent integer pels. This permits determination of the coefficients of the quadratic model. An estimated cost is obtained using the actual fractional pel position in the quadratic model. This can be used for obtaining the cost of a P Skip mode. This can be used for the decision of B direct or B skip modes in B frames.

Abstract: Methods for processing of video sequences that may contain telecined (3:2 pull down) frame sequences are provided. A method for detecting 3:2 pull down is provided that measures vertical detail in frames of a video sequence and uses the variation in vertical detail over time to decide whether the video sequence contains normal interlace content or 3:2 pull down content. A method for improving the compression of detected 3:2 pull down content is also provided that controls the selection of field or frame coding mode for frames of 3:2 pull down content and the selection of reference fields for encoding duplicated fields in the 3:2 pull down content.

Abstract: This invention is method for determining direction mode for an image block in intra prediction of video frames. The method calculates a plurality of energy terms from the spatial frequency domain image block, then determines a coarse classification from these energy terms. For at least some coarse classifications the method searches from among a subset of less than all possible direction modes. The method also searches a most likely direction mode determined from neighbor blocks when available.

Abstract: A method of video coding comprising Intra prediction coding at least a predetermined constant M macroblocks within an ordinary frame. These M macroblocks include naturally selected macroblocks where the bit cost of Intra prediction coding is less than for Inter prediction coding and optional additional macroblocks. The optional additional macroblocks have the least bit cost of Inter prediction coding as adjusted by a penalty factor each time a corresponding macroblock is Intra coded in a predetermined number of frames. An end frame Intra prediction codes each macroblock having no corresponding Intra prediction coded macroblock in a predetermined number of ordinary frames.

Abstract: This invention enables determination of the best video encoding mode for a macroblock in the SKIP mode and comparing this cost with other modes. This invention avoids sub-pixel interpolations for fractional pixels. This invention models the cost function in a quadratic model and computes the cost for a nearest integer pel and plural adjacent integer pels. This permits determination of the coefficients of the quadratic model. An estimated cost is obtained using the actual fractional pel position in the quadratic model. This method can be used for obtaining the cost of a P Skip mode. This method can be used for the decision of B direct or B skip modes in B frames.

Abstract: This invention is method for determining direction mode for an image block in intra prediction of video frames. The method calculates a plurality of energy terms from the spatial frequency domain image block, then determines a coarse classification from these energy terms. For at least some coarse classifications the method searches from among a subset of less than all possible direction modes. The method also searches a most likely direction mode determined from neighbor blocks when available.

Abstract: In one aspect, code-words of variable lengths are decoded using a multi-stage decoding approach, with different stages being of different sizes (and thus accepting input data of corresponding number of bits). According to another aspect, the same bit positions are used for storing symbol information and offset value in case of hit and miss results respectively, thereby reducing the width of the entries of the decoding table. According to yet another aspect, conditional processing is avoided by providing a common arithmetic operation when decoding code-words in various escape modes, but using neutral operand values in case the operation is not required.

Abstract: In one aspect, code-words of variable lengths are decoded using a multi-stage decoding approach, with different stages being of different sizes (and thus accepting input data of corresponding number of bits). According to another aspect, the same bit positions are used for storing symbol information and offset value in case of hit and miss results respectively, thereby reducing the width of the entries of the decoding table. According to yet another aspect, conditional processing is avoided by providing a common arithmetic operation when decoding code-words in various escape modes, but using neutral operand values in case the operation is not required.