Patents Assigned to HFI INNOVATIONS INC.
-
Patent number: 11997311Abstract: Video processing methods and apparatuses for coding a current block generate a final predictor by combining multiple predictors of the current block. One of the predictors for the current block is a motion compensated predictor, which is generated according to one candidate selected from a first candidate list. The construction of the first candidate list includes deriving an average candidate by averaging motion information of existing candidates in the first candidate list. A second predictor for the current block is another motion compensated predictor or an intra predictor.Type: GrantFiled: September 17, 2019Date of Patent: May 28, 2024Assignee: HFI INNOVATION INC.Inventors: Man-Shu Chiang, Chih-Wei Hsu
-
Patent number: 11997266Abstract: Methods and apparatus for loop-filter processing of reconstructed video are disclosed. According to one method, multiple CC-ALF (Cross-Component Adaptive Loop Filter) filters are used. Selection of the multiple CC-ALF filters can be signalled in one APS (Adaptation Parameter Set). According to another method, the CC-ALF can be implemented according to the difference between a to-be-process sample and its neighbouring sample.Type: GrantFiled: July 24, 2020Date of Patent: May 28, 2024Assignee: HFI INNOVATION INC.Inventors: Yu-Wen Huang, Chih-Wei Hsu, Ching-Yeh Chen, Tzu-Der Chuang, Shaw-Min Lei
-
Patent number: 11991393Abstract: Video processing methods and apparatuses in a video encoding or decoding system for transforming residuals of transform blocks into final transform coefficients or inverse transforming final transform coefficients into residuals. In order to solve the latency issue, exemplary embodiments check if a width or height of a coding block is larger than a predefined threshold, and disable secondary transform or inverse secondary transform for any transform block within the coding block if the width or height of the coding block is larger than the predefined threshold. Another embodiment checks if there are multiple transform blocks in a coding block, and disables secondary transform or inverse secondary transform if the coding block contains multiple transform blocks.Type: GrantFiled: April 15, 2020Date of Patent: May 21, 2024Assignee: HFI INNOVATION INC.Inventors: Man-Shu Chiang, Chih-Wei Hsu, Tzu-Der Chuang, Ching-Yeh Chen
-
Patent number: 11985330Abstract: Methods and apparatus of for video coding using sub-block based affine mode are disclosed. In one method, if affine fallback is used or the control-point motion vectors are the same, the sub-block based affine mode is disabled in order to reduce computational complexity. According to another method for video coding using a coding tool belonging to a coding tool group comprising Prediction Refinement with Optical Flow (PROF) and Bi-Directional Optical Flow (BDOF), predictor refinement is derived for pixels of a target subblock of the current block, where a step to derive the predictor refinement comprises to derive gradients for the pixels of the target subblock of the current block and to right-shift the first gradients by a common shift.Type: GrantFiled: April 10, 2020Date of Patent: May 14, 2024Assignee: HFI INNOVATION INC.Inventors: Chen-Yen Lai, Tzu-Der Chuang, Ching-Yeh Chen
-
Patent number: 11985314Abstract: Video processing methods and apparatuses in a video encoding or decoding system for processing out-of-bounds nodes in a current picture. An out-of-bounds node is a coding tree node with a block region across a current picture boundary. The video processing method or apparatus determines an inferred splitting type, applies the inferred splitting type to split the out-of-bounds node into child blocks, adaptively splits each child block into one or multiple leaf blocks, and encodes or decodes the leaf blocks in the out-of-bounds node inside the current picture. The inferred splitting type for partitioning out-of-bounds nodes in an inter slice, picture, or tile is the same as the inferred splitting type for partitioning out-of-bounds nodes in an intra slice, picture, or tile.Type: GrantFiled: December 24, 2019Date of Patent: May 14, 2024Assignee: HFI INNOVATION INC.Inventors: Chia-Ming Tsai, Chih-Wei Hsu, Tzu-Der Chuang, Ching-Yeh Chen, Yu-Wen Huang, Shih-Ta Hsiang
-
Patent number: 11985324Abstract: Exemplary video processing methods and apparatuses for encoding or decoding a current block by inter prediction are disclosed. Input data of a current block is received and partitioned into sub-partitions and motion refinement is independently performed on each sub-partition. A reference block for each sub-partition is obtained from one or more reference pictures according to an initial motion vector (MV). A refined MV for each sub-partition is derived by searching around the initial MV with N-pixel refinement. One or more boundary pixels of the reference block for a sub-partition is padded for motion compensation of the sub-partition. A final predictor for the current block is generated by performing motion compensation for each sub-partition according to its refined MV. The current block is then encoded or decoded according to the final predictor.Type: GrantFiled: March 13, 2020Date of Patent: May 14, 2024Assignee: HFI INNOVATION INC.Inventors: Yu-Cheng Lin, Chun-Chia Chen, Chih-Wei Hsu, Ching-Yeh Chen, Tzu-Der Chuang, Yu-Wen Huang
-
Patent number: 11979613Abstract: Encoding methods and apparatuses include receiving input video data of a current block in a current picture and applying a Cross-Component Adaptive Loop Filter (CCALF) processing on the current block based on cross-component filter coefficients to refine chroma components of the current block according to luma sample values. The method further includes signaling two Adaptive Loop Filter (ALF) signal flags and two CCALF signal flags in an Adaptation Parameter Set (APS) with an APS parameter type equal to ALF or parsing two ALF signal flags and two CCALF signal flags from an APS with an APS parameter type equal to ALF, signaling or parsing one or more Picture Header (PH) CCALF syntax elements or Slice Header (SH) CCALF syntax elements, wherein both ALF and CCALF signaling are present either in a PH or SH, and encoding or decoding the current block in the current picture.Type: GrantFiled: June 28, 2022Date of Patent: May 7, 2024Assignee: HFI INNOVATION INC.Inventors: Ching-Yeh Chen, Olena Chubach, Chen-Yen Lai, Tzu-Der Chuang, Chih-Wei Hsu, Yu-Wen Huang
-
Patent number: 11973958Abstract: Methods and apparatus of video coding using sub-block based affine mode are disclosed. According to this method, control-point motion vectors (MVs) associated with the affine mode are determined for a block. A sub-block MV is derived for a target sub-block of the block from the control-point MVs for the block. A prediction offset is determined for a target pixel of the target sub-block using information comprising a pixel MV offset from the sub-block MV for the target pixel according to Prediction Refinement with Optical Flow (PROF). The target pixel of the target sub-block is encoded or decoded using a modified predictor. The modified prediction is generated by clipping the prediction offset to a target range and combining the clipped prediction offset with an original predictor.Type: GrantFiled: September 22, 2020Date of Patent: April 30, 2024Assignee: HFI INNOVATION INC.Inventors: Tzu-Der Chuang, Ching-Yeh Chen, Zhi-Yi Lin
-
Patent number: 11956469Abstract: Video processing methods and apparatuses implemented in a video encoding or decoding system with conditional secondary transform signaling. The video encoding system determines and applies a transform operation to residuals of a transform block to generate final transform coefficients, and adaptively signals a secondary transform index according to a position of a last significant coefficient in the transform block. A value of the secondary transform index is determined according to the transform operation.Type: GrantFiled: April 15, 2020Date of Patent: April 9, 2024Assignee: HFI INNOVATION INC.Inventors: Man-Shu Chiang, Chih-Wei Hsu, Tzu-Der Chuang, Ching-Yeh Chen
-
Patent number: 11956421Abstract: Method and apparatus of video coding are disclosed. According to one method, in the decoder side, a predefined Intra mode is assigned to a neighboring block adjacent to the current luma block when the neighboring block satisfies one or more conditions. An MPM (Most Probable Mode) list is derived based on information comprising at least one of neighboring Intra modes. A current Intra mode is derived utilizing the MPM list. The current luma block is decoded according to the current Intra mode According to another method, a predefined Intra mode is assigned to a neighboring block adjacent to the current luma block if the neighboring block is coded in BDPCM (Block-based Delta Pulse Code Modulation) mode, where the predefined Intra mode is set to horizontal mode or vertical mode depending on prediction direction used by the BDPCM mode.Type: GrantFiled: May 7, 2020Date of Patent: April 9, 2024Assignee: HFI INNOVATION INC.Inventors: Man-Shu Chiang, Chih-Wei Hsu, Tzu-Der Chuang, Ching-Yeh Chen, Yu-Wen Huang, Shih-Ta Hsiang
-
Patent number: 11956462Abstract: Video processing methods and apparatuses for coding a current block comprise receiving input data of a current block, partitioning the current block into multiple sub-blocks, deriving sub-block MVs for the current block according to a sub-block motion compensation coding tool, constraining the sub-block MVs to form constrained sub-block MVs, and encoding or decoding the current block using the constrained sub-block MVs, and applying motion compensation to the current block using the constrained sub-block MVs to encode or decode the current block. The sub-block MVs may be constrained according to a size, width, or height of the current block or a sub-block, an inter prediction direction of one of control point MVs of the current block, the current block, or current sub-block, the control point MVs, or a combination of the above.Type: GrantFiled: December 8, 2021Date of Patent: April 9, 2024Assignee: HFI INNOVATION INC.Inventors: Tzu-Der Chuang, Ching-Yeh Chen, Chen-Yen Lai, Chih-Wei Hsu
-
Patent number: 11949852Abstract: A method and apparatus of video coding, where according to one method, input data related to a current block in a current picture are received at a video encoder side or compressed data comprising the current block are received at a video decoder side. A first syntax at a high level in a video bitstream regarding residual coding type is signaled at the encoder side or parsed at the decoder side. A target coding mode is determined for the current block based on information comprising a value of the first syntax. The current block is encoded at the encoder side or decoded at the decoder side according to the target coding mode. The high level may correspond to a slice header or a picture header.Type: GrantFiled: September 16, 2020Date of Patent: April 2, 2024Assignee: HFI INNOVATION INC.Inventors: Zhi-Yi Lin, Tzu-Der Chuang, Ching-Yeh Chen
-
Patent number: 11943476Abstract: Video processing methods and apparatuses implemented in a video encoding or decoding system with conditional secondary transform signaling. The video encoding system determines and applies a transform operation to residuals of one or more transform blocks to generate final transform coefficients, and skip signaling a secondary transform index if a position of a last significant coefficient in each considered transform block is less than or equal to a predefined position; otherwise, the video encoding system signals a secondary transform index according to the transform operation.Type: GrantFiled: April 15, 2020Date of Patent: March 26, 2024Assignee: HFI INNOVATION INC.Inventors: Man-Shu Chiang, Chih-Wei Hsu, Tzu-Der Chuang, Ching-Yeh Chen
-
Patent number: 11930169Abstract: Methods and apparatus for video coding are disclosed. According to one method, First ALF (Adaptive Loop Filter) processing is applied to the reconstructed chroma samples for a target reconstructed chroma sample to generate a first filtered chroma sample. Second ALF processing is applied to the related reconstructed luma samples to generate a second filtered chroma sample for the target reconstructed chroma sample, where positions of the related reconstructed luma samples selected for the second ALF processing are determined according to the target chroma format. According to another method, the luma ALF and the cross-component ALF have the same filter coefficient precision.Type: GrantFiled: June 24, 2020Date of Patent: March 12, 2024Assignee: HFI INNOVATION INC.Inventors: Chen-Yen Lai, Ching-Yeh Chen, Tzu-Der Chuang
-
Patent number: 11930174Abstract: A method and apparatus for block partition are disclosed. If a cross-colour component prediction mode is allowed, the luma block and the chroma block are partitioned into one or more luma leaf blocks and chroma leaf blocks. If a cross-colour component prediction mode is allowed, whether to enable an LM (Linear Model) mode for a target chroma leaf block is determined based on a first split type applied to an ancestor chroma node of the target chroma leaf block and a second split type applied to a corresponding ancestor luma node. According to another method, after the luma block and the chroma block are partitioned using different partition tress, determine whether one or more exception conditions to allow an LM for a target chroma leaf block are satisfied when the chroma partition tree uses a different split type, a different partition direction, or both from the luma partition tree.Type: GrantFiled: December 30, 2019Date of Patent: March 12, 2024Assignee: HFI INNOVATION INC.Inventors: Chia-Ming Tsai, Tzu-Der Chuang, Chih-Wei Hsu, Ching-Yeh Chen, Zhi-Yi Lin
-
Publication number: 20240080490Abstract: A video codec receives data for a block of pixels to be encoded or decoded as a current block of a current picture of a video. The video codec signals or parses a first syntax element for a first coding mode in a particular set of two or more coding modes. Each of coding mode of the particular set of coding modes modifies a merge candidate or an inter-prediction that is generated based on the merge candidate. The video codec enables the first coding mode and disables one or more other coding modes in the particular set of coding modes. The disabled one or more coding modes in the particular set of coding modes are disabled without parsing syntax elements for the disabled coding modes. The video codec encodes or decodes the current block by using the enabled first coding mode and bypassing the disabled coding modes.Type: ApplicationFiled: November 10, 2023Publication date: March 7, 2024Applicant: HFI Innovation Inc.Inventors: Man-Shu Chiang, Chih-Wei Hsu, Ching-Yeh Chen
-
Patent number: 11924426Abstract: A video system that applies constraints on block partitioning is provided. The system receives a partitioning control parameter from a bitstream specifying a maximum block size for enabling ternary-tree split that is constrained to be 64 or smaller. The system receives data from a bitstream for a block of pixels to be decoded as a current block of a current picture of a video. The system splits the current block into one or more partitions recursively, wherein ternary split is disallowed for a partition of the current block unless the partition is less than or equal to the maximum block size. The system reconstructs the one or more partitions of the current block.Type: GrantFiled: January 23, 2023Date of Patent: March 5, 2024Assignee: HFI INNOVATION INC.Inventors: Shih-Ta Hsiang, Chen-Yen Lai, Ching-Yeh Chen
-
Patent number: 11924444Abstract: Method and apparatus for constrained de-blocking filter are disclosed. One method receives input data related to a current block in a current picture at a video encoder side or a video bitstream corresponding to compressed data including the current block in the current picture at a video decoder side, and determines a first boundary associated with the current block, wherein the first boundary corresponds to one vertical boundary or one horizontal boundary of the current block. The method then applies de-blocking process to a reconstructed current block corresponding to the current block to result in a filtered-reconstructed current block, using a plurality of first reference samples at a same side of the first boundary, and replaces a first set of the first reference samples by one or more padding values. The method then generates a filtered decoded picture including the filtered-reconstructed current block.Type: GrantFiled: April 11, 2022Date of Patent: March 5, 2024Assignee: HFI INNOVATION INC.Inventors: Chia-Ming Tsai, Chih-Wei Hsu, Ching-Yeh Chen, Tzu-Der Chuang, Yu-Wen Huang
-
Patent number: 11917185Abstract: A method and apparatus of Inter prediction for video coding using Multi-hypothesis (MH) are disclosed. If an MH mode is used for the current block: at least one MH candidate is derived using reduced reference data by adjusting at least one coding-control setting; an Inter candidate list is generated, where the Inter candidate list comprises said at least one MH candidate; and current motion information associated with the current block is encoded using the Inter candidate list at the video encoder side or the current motion information associated with the current block is decoded at the video decoder side using the Merge candidate list. The coding control setting may correspond to prediction direction setting, filter tap setting, block size of reference block to be fetched, reference picture setting or motion limitation setting.Type: GrantFiled: November 30, 2021Date of Patent: February 27, 2024Assignee: HFI INNOVATION INC.Inventors: Man-Shu Chiang, Chih-Wei Hsu, Tzu-Der Chuang, Ching-Yeh Chen, Yu-Wen Huang
-
Patent number: 11909965Abstract: Methods and apparatus for Adaptive Loop Filter (ALF) processing of reconstructed video are disclosed. According to one implementation, a method receives reconstructed pixels of a current block and determines one or more clipping values for the ALF processing, wherein a target clipping value is always signaled at an encoder side or parsed at a decoder side even if the target clipping value is zero. The method then derives a current ALF output for the current block, wherein the current ALF output has a weighted sum of clipped differences and each of the original differences is calculated between a first reconstructed pixel at a non-center filter location and a center reconstructed pixel at a center filter location, and each of the original differences is clipped according to a corresponding clipping value to form one clipped difference. The method then provides filtered-reconstructed pixels.Type: GrantFiled: August 12, 2022Date of Patent: February 20, 2024Assignee: HFI INNOVATION INC.Inventors: Chen-Yen Lai, Ching-Yeh Chen, Tzu-Der Chuang, Yu-Wen Huang