Abstract: Encoders and methods of encoding that incorporate deblocking into the encoding process. An encoder performs a two-cycle encoding process. First, an original block is processed and then reconstructed and deblocked. The reconstruction and the deblocked reconstruction are compared to the original and an intermediate block is created that contains the portions of the reconstruction or the deblocked reconstruction that are more similar to the original. In the second cycle, the intermediate block serves to generate a modified original block, which is then used in a prediction process to create a new prediction block. The new prediction block as compared to the original gives new residuals. The new prediction and new residuals are entropy coded to generate the encoded video data. The new prediction block and new residuals incorporate the deblocking elements selectively identified during the first cycle, thereby eliminating the need for deblocking at the decoder.

Abstract: A method of encoding a video using constrained soft-decision quantization. The soft-decision quantization includes first performing hard-decision quantization to obtain hard quantized coefficients and, then, obtaining a soft quantized coefficient using a rate-distortion calculation over a search range of quantization levels for a transform domain coefficient, wherein the search range of quantization levels for that transform domain coefficient is constrained within a number of quantization levels of a corresponding hard quantized coefficient. The search range may be based upon a fixed threshold, the coefficient position, the hard quantized coefficient magnitude, a threshold value less accumulated distortion, or other factors, including combinations of these factors. The accumulated distortion may be measured by an L1 norm.

Abstract: Methods of encoding and decoding for video data are describe in which significance maps are encoded and decoded using non-spatially-uniform partitioning of the map into parts, wherein the bit positions within each part are associated with a given context. Example partition sets and processes for selecting from amongst predetermined partition sets and communicating the selection to the decoder are described.

Abstract: Digital video content is processed for delivery over a communications channel by segmenting the digital video content into one or more regions of interest (ROI) in accordance with content signature of the video content and encoding the digital video content in accordance with the ROI segmentation and the communications channel.

Abstract: Encoders and methods of encoding that incorporate deblocking into the encoding process. An encoder performs a two-cycle encoding process. First, an original block is processed and then reconstructed and deblocked. The reconstruction and the deblocked reconstruction are compared to the original and an intermediate block is created that contains the portions of the reconstruction or the deblocked reconstruction that are more similar to the original. In the second cycle, the intermediate block serves to generate a modified original block, which is then used in a prediction process to create a new prediction block. The new prediction block as compared to the original gives new residuals. The new prediction and new residuals are entropy coded to generate the encoded video data. The new prediction block and new residuals incorporate the deblocking elements selectively identified during the first cycle, thereby eliminating the need for deblocking at the decoder.

Abstract: A method of encoding a video using constrained soft-decision quantization. The soft-decision quantization includes first performing hard-decision quantization to obtain hard quantized coefficients and, then, obtaining a soft quantized coefficient using a rate-distortion calculation over a search range of quantization levels for a transform domain coefficient, wherein the search range of quantization levels for that transform domain coefficient is constrained within a number of quantization levels of a corresponding hard quantized coefficient. The search range may be based upon a fixed threshold, the coefficient position, the hard quantized coefficient magnitude, a threshold value less accumulated distortion, or other factors, including combinations of these factors. The accumulated distortion may be measured by an L1 norm.

Abstract: A video encoder sends at least some information regarding boundary strength to the decoder along with the bitstream of encoded video. The decoder is configured to use the received boundary strength information from the encoder to reduce the number of computations necessary for the decoder to determine the boundary strength details required for performing deblocking when decoding the bitstream.

Abstract: Digital video content is processed for delivery over a communications channel by segmenting the digital video content into one or more regions of interest (ROI) in accordance with content signature of the video content and encoding the digital video content in accordance with the ROI segmentation and the communications channel.