# Xiang Yu, Waterloo CA

## Xiang Yu, Waterloo CA

Patent application number | Description | Published |
---|---|---|

20080298699 | METHOD AND DEVICE FOR DOWN-SAMPLING A DCT IMAGE IN THE DCT DOMAIN - A method and device that uses transform matrices to down-sample a DCT image directly in the DCT domain. The transform matrices have been selected to minimize an optimization problem which is a function of the visual quality of down-sampled images obtained using the matrices and the computational complexity associated with using the transform matrices. The transform matrices comprise a row transform matrix and a column transform matrix. A down-sampled image is produced by determining an intermediary matrix as the product of the DCT image and one of either the row transform matrix or the column transform matrix and then determining the down-sampled image as the product of the intermediary matrix and the transform matrix not already used to determine the intermediary matrix. | 12-04-2008 |

20080298700 | METHOD AND DEVICE FOR SELECTING TRANSFORM MATRICES FOR DOWN-SAMPLING DCT IMAGE USING LEARNING WITH FORGETTING ALGORITHM - Down-sampling of an image may be performed in the DCT domain. A multiple layered network is used to select transform matrices for down-sampling a DCT image of size M×N to a DCT image of size I×J. A spatial domain down-sampling method is selected and applied to the DCT image to produce a down-sampled DCT reference image. A learning with forgetting algorithm is used to apply a decay to the elements of the transform matrix and select a transform matrices which solve an optimization problem. The optimization problem is a function of the visual quality of images obtained using the transform matrices and the computational complexity associated with using the transform matrices. The visual quality is a measure of the difference between the down-sampled DCT image obtained using the transform matrices and the visual quality of the DCT reference image obtained using a spatial domain down-sampling method. | 12-04-2008 |

20080298701 | Method and Device for Selecting Optimal Transform Matrices for Down-sampling DCT image - Down-sampling of an image may be performed in the DCT domain. Transform matrices are obtained for down-sampling a DCT image of size M×N to a down-sampled DCT image of size I×J. The transform matrices may be used to down-sample the DCT image directly in the DCT domain. A spatial domain down-sampling method is selected and applied to the DCT image to produce a down-sampled DCT reference image. The transform matrices are selected by solving an optimization problem, leading to transform matrices which achieve a desired trade-off between the visual quality of images obtained using the transform matrices and the computational complexity associated with using the transform matrices. The visual quality is a measure of the difference between the down-sampled DCT image obtained using the transform matrices and the visual quality of the DCT reference image obtained using a spatial domain down-sampling method. | 12-04-2008 |

20090207937 | Method and System for Optimizing Quantization for Noisy Channels - Methods are described for configuring a quantizer to achieve improved end-to-end distortion performance when transmitting encoded source data over a noisy channel. The codebook and partitioning are selected using an iterative process of determining an updated codebook and an updated partition space, where the updated codebook is based, in part, on the average symbol error probability of the channel. Complete knowledge of the transitional probabilities of the channel is not required. Variants of the iterative process are described. | 08-20-2009 |

20110019737 | SOFT DECISION AND ITERATIVE VIDEO CODING FOR MPEG AND H.264 - A method of encoding video data using soft decision quantization makes use of iterative encoding to provide the ability to optimize encoding across different functional elements in a hybrid video encoder. Iterative encoding is used to allow quantization step size, motion prediction and quantization levels to be optimized despite their interrelated nature. | 01-27-2011 |

20110184733 | SYSTEM AND METHOD FOR ENCODING AND DECODING PULSE INDICES - Methods, and corresponding codec-containing devices are provided that have source coding schemes for encoding a component of an excitation. In some cases, the source coding scheme is an enumerative source coding scheme, while in other cases the source coding scheme is an arithmetic source coding scheme. In some cases, the source coding schemes are applied to encode a fixed codebook component of the excitation for a codec employing codebook excited linear prediction, for example an AMR-WB (Adaptive Multi-Rate-Wideband) speech codec. | 07-28-2011 |

20110235932 | METHOD AND DEVICE FOR SELECTING OPTIMAL TRANSFORM MATRICES FOR DOWN-SAMPLING DCT IMAGE - Down-sampling of an image may be performed in the DCT domain. Transform matrices are obtained for down-sampling a DCT image of size M×N to a down-sampled DCT image of size I×J. The transform matrices may be used to down-sample the DCT image directly in the DCT domain. A spatial domain down-sampling method is selected and applied to the DCT image to produce a down-sampled DCT reference image. The transform matrices are selected by solving an optimization problem, leading to transform matrices which achieve a desired trade-off between the visual quality of images obtained using the transform matrices and the computational complexity associated with using the transform matrices. The visual quality is a measure of the difference between the down-sampled DCT image obtained using the transform matrices and the visual quality of the DCT reference image obtained using a spatial domain down-sampling method. | 09-29-2011 |

20110249722 | METHOD AND DEVICE FOR SELECTING TRANSFORM MATRICES FOR DOWN-SAMPLING DCT IMAGE USING LEARNING WITH FORGETTING ALGORITHM - Down-sampling of an image may be performed in the DCT domain. A multiple layered network is used to select transform matrices for down-sampling a DCT image of size M×N to a DCT image of size I×J. A spatial domain down-sampling method is selected and applied to the DCT image to produce a down-sampled DCT reference image. A learning with forgetting algorithm is used to apply a decay to the elements of the transform matrix and select a transform matrices which solve an optimization problem. The optimization problem is a function of the visual quality of images obtained using the transform matrices and the computational complexity associated with using the transform matrices. The visual quality is a measure of the difference between the down-sampled DCT image obtained using the transform matrices and the visual quality of the DCT reference image obtained using a spatial domain down-sampling method. | 10-13-2011 |

20120014435 | Soft Decision and Iterative Video Coding for MPEG and H.264 - A method of encoding video data using soft decision quantization makes use of iterative encoding to provide the ability to optimize encoding across different functional elements in a hybrid video encoder. Iterative encoding is used to allow quantization step size, motion prediction and quantization levels to be optimized despite their interrelated nature. | 01-19-2012 |

20120093428 | METHOD AND DEVICE FOR DOWN-SAMPLING A DCT IMAGE IN THE DCT DOMAIN - A method and device that uses transform matrices to down-sample a DCT image directly in the DCT domain. The transform matrices have been selected to minimize an optimization problem which is a function of the visual quality of down-sampled images obtained using the matrices and the computational complexity associated with using the transform matrices. The transform matrices comprise a row transform matrix and a column transform matrix. A down-sampled image is produced by determining an intermediary matrix as the product of the DCT image and one of either the row transform matrix or the column transform matrix and then determining the down-sampled image as the product of the intermediary matrix and the transform matrix not already used to determine the intermediary matrix. | 04-19-2012 |

20120219055 | METHODS AND DEVICES FOR DATA COMPRESSION USING OFFSET-BASED ADAPTIVE RECONSTRUCTION LEVELS - Encoding and decoding methods are presented that used offset-based adaptive reconstruction levels. The offset data is inserted in the bitstream with the encoded video data. The offset data may be differential data and may be an index to an array of offset values from which the differential offset is calculated by the decoder. The offset to an adaptive reconstruction level may be adjusted for each slice. The offsets may be specific to a particular level/index and data type. In some cases, offsets may only be sent for a subset of the levels. Higher levels may apply no offset, may apply an average offset, or may apply the offset used for the highest level having a level-specific offset. | 08-30-2012 |

20120263375 | METHOD AND DEVICE FOR SELECTING OPTIMAL TRANSFORM MATRICES FOR DOWN-SAMPLING DCT IMAGE - Down-sampling of an image may be performed in the DCT domain. Transform matrices are obtained for down-sampling a DCT image of size M×N to a down-sampled DCT image of size I×J. The transform matrices may be used to down-sample the DCT image directly in the DCT domain. A spatial domain down-sampling method is selected and applied to the DCT image to produce a down-sampled DCT reference image. The transform matrices are selected by solving an optimization problem, leading to transform matrices which achieve a desired trade-off between the visual quality of images obtained using the transform matrices and the computational complexity associated with using the transform matrices. The visual quality is a measure of the difference between the down-sampled DCT image obtained using the transform matrices and the visual quality of the DCT reference image obtained using a spatial domain down-sampling method. | 10-18-2012 |

20140307781 | METHODS AND DEVICES FOR DATA COMPRESSION USING OFFSET-BASED ADAPTIVE RECONSTRUCTION LEVELS - Encoding and decoding methods are presented that used offset-based adaptive reconstruction levels. The offset data is inserted in the bitstream with the encoded video data. The offset data may be differential data and may be an index to an array of offset values from which the differential offset is calculated by the decoder. The offset to an adaptive reconstruction level may be adjusted for each slice. The offsets may be specific to a particular level/index and data type. In some cases, offsets may only be sent for a subset of the levels. Higher levels may apply no offset, may apply an average offset, or may apply the offset used for the highest level having a level-specific offset. | 10-16-2014 |

20140376622 | METHODS AND DEVICES FOR DATA COMPRESSION USING A NON-UNIFORM RECONSTRUCTION SPACE - An encoding method for encoding video data by adjusting a quantization parameter, the video data being partitioned into blocks comprising sets of quantized transform coefficients. The method includes, for a set of quantized transform coefficients corresponding to one of the blocks, collecting statistics, wherein the statistics comprise the number of quantized transform coefficients and the sum of the non-rounded quantization value of the quantized transform coefficients in the set. The method also includes deriving a step size based on the statistics, mapping the derived step size to a closest quantization parameter value, and quantizing a next block using the mapped quantization parameter value. | 12-25-2014 |