Patents by Inventor Marta Karczewicz

Marta Karczewicz has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20250047884
    Abstract: A video decoder may be configured to receive a block of video data that was encoded using a coding mode that includes a search process in one or more reference frames. The video decoder may prefetch reference samples in a fixed search region of at least one reference frame of the one or more reference frames, and decode the block of video data using the coding mode, including performing the search process for the coding mode using the prefetched reference samples.
    Type: Application
    Filed: August 4, 2023
    Publication date: February 6, 2025
    Inventors: Cheng-Teh Hsieh, Han Huang, Chun-Chi Chen, Marta Karczewicz
  • Publication number: 20250039433
    Abstract: An example device for coding video data includes a memory configured to store video data; and one or more processing units implemented in circuitry and configured to: store motion information for a first coding tree unit (CTU) line of a picture in a first history motion vector predictor (MVP) buffer of the memory; reset a second history MVP buffer of the memory; and after resetting the second history MVP buffer, store motion information for a second CTU line of the picture in the second history MVP buffer, the second CTU line being different than the first CTU line. Separate threads of a video coding process executed by the one or more processors may process respective CTU lines, in some examples.
    Type: Application
    Filed: July 5, 2024
    Publication date: January 30, 2025
    Inventors: Luong Pham Van, Wei-Jung Chien, Vadim Seregin, Marta Karczewicz, Han Huang
  • Patent number: 12212736
    Abstract: A device for decoding video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: determine a deterministic bounding box from which to retrieve reference samples of reference pictures of video data for performing decoder-side motion vector derivation (DMVD) for a current block of the video data; derive a motion vector for the current block according to DMVD using the reference samples within the deterministic bounding box; form a prediction block using the motion vector; and decode the current block using the prediction block.
    Type: Grant
    Filed: June 27, 2022
    Date of Patent: January 28, 2025
    Assignee: QUALCOMM INCORPORATED
    Inventors: Chun-Chi Chen, Han Huang, Cheng-Teh Hsieh, Wei-Jung Chien, Zhi Zhang, Yao-Jen Chang, Yan Zhang, Vadim Seregin, Marta Karczewicz
  • Publication number: 20250030870
    Abstract: A method of encoding or decoding video data includes for each of a plurality of models, deriving respective model parameters associated with respective corresponding models of the plurality of models utilizing a first set of samples; determining a respective cost value associated with the corresponding models of the plurality of models utilizing the respective model parameters and a second set of samples to generate respective cost values for the plurality of models, wherein the first set of samples and the second set of samples are exclusive of one another; determining a model of the plurality of models to use for encoding or decoding a current block based on the respective cost values; and encoding or decoding the current block based on the determined model.
    Type: Application
    Filed: July 15, 2024
    Publication date: January 23, 2025
    Inventors: Vadim Seregin, Marta Karczewicz
  • Publication number: 20250030877
    Abstract: Example devices and techniques are described. An example device includes one or more memories configured to store video data and one or more processors configured to receive, in a bitstream, one or more first messages each specifying at least one respective film grain characteristic applicable to first video data of the video data and each corresponding to a different film grain synthesis technique. The one or more processors are configured to determine that a first film grain synthesis technique specified by a first message of the one or more first messages corresponds to a device supported film grain synthesis technique. The one or more processors are configured to apply, based on a determination that the first film grain synthesis technique corresponds to a device supported film grain synthesis technique, the device supported film grain synthesis technique to the first video data based on contents of the first message.
    Type: Application
    Filed: June 26, 2024
    Publication date: January 23, 2025
    Inventors: Vadim Seregin, Marta Karczewicz, Muhammed Zeyd Coban
  • Publication number: 20250030888
    Abstract: Techniques are described herein for processing video data using enhanced interpolation filters for intra-prediction. For instance, a device can determine an intra-prediction mode for predicting a block of video data. The device can determine a type of smoothing filter to use for the block of video data, wherein the type of the smoothing filter is determined based at least in part on comparing at least one of a width of the block of video data and a height of the block of video data to a first threshold. The device can further perform intra-prediction for the block of video data using the determined type of smoothing filter and the intra-prediction mode.
    Type: Application
    Filed: October 7, 2024
    Publication date: January 23, 2025
    Inventors: Bappaditya RAY, Vadim SEREGIN, Marta KARCZEWICZ
  • Patent number: 12206864
    Abstract: A method of decoding video data including, receiving a first block of video data, receiving a first syntax element indicating if a coding mode is to be used for the first block of video data in the case that the first block of video data is associated with a number of non-zero transform coefficients greater than or equal to a threshold, explicitly decoding a value of the received first syntax element, and applying the coding mode to the first block of video data in accordance with a value of the first syntax element.
    Type: Grant
    Filed: September 20, 2019
    Date of Patent: January 21, 2025
    Assignee: QUALCOMM INCORPORATED
    Inventors: Vadim Seregin, Marta Karczewicz, Xin Zhao
  • Patent number: 12200207
    Abstract: A video decoder is configured to determine, for a current block of video data, a classifier from a plurality of available classifiers for the current block, wherein each classifier of the plurality of available classifiers defines a process for calculating a class index; determine a class index based on the classifier and one or more sample values of the current block; select a filter from a set of filters based on the class index; and apply the filter to the current block.
    Type: Grant
    Filed: June 29, 2022
    Date of Patent: January 14, 2025
    Assignee: QUALCOMM Incorporated
    Inventors: Nan Hu, Vadim Seregin, Marta Karczewicz
  • Patent number: 12200219
    Abstract: An example method of encoding or decoding video data includes determining a motion vector for a block of video data using a decoder side motion vector derivation process that includes performing an iterative search process, wherein performing the iterative search process includes: determining a minimum cost through a previous search iteration; updating an overall minimum cost through a current search iteration; and terminating the iterative search process early based on a comparison of the minimum cost through the previous search iteration and the overall minimum cost through the current search iteration; and encoding or decoding the block of video data using the motion vector.
    Type: Grant
    Filed: September 6, 2022
    Date of Patent: January 14, 2025
    Assignee: QUALCOMM Incorporated
    Inventors: Vadim Seregin, Han Huang, Marta Karczewicz
  • Publication number: 20250016339
    Abstract: An example device for decoding video data includes a processing system comprising one or more processors implemented in circuitry and configured to: determine that a first temporal layer identifier of a first picture of the video data is included in a first set of temporal layers; in response to the first temporal layer identifier being included in the first set of temporal layers, decode blocks of the first picture on a block by block basis; determine that a second temporal layer identifier of a second picture of the video data is included in a second set of temporal layers, the second set of temporal layers being higher than the first set of temporal layers; and in response to the second temporal layer identifier being included in the second set of temporal layers, execute a neural network-based video decoder to decode the second picture.
    Type: Application
    Filed: June 14, 2024
    Publication date: January 9, 2025
    Inventors: Thomas Alexander Ryder, Samuel James Eadie, Marta Karczewicz, Muhammed Zeyd Coban, Vadim Seregin
  • Publication number: 20250016326
    Abstract: A video decoder can be configured to determine that a current block of the video data is coded in a bi-prediction inter mode; receive a first syntax element identifying a motion vector predictor from a first candidate list of motion vector predictors; receive a second syntax element identifying a motion vector difference; determine a first motion vector for the current block based on the motion vector predictor and the motion vector difference; determine a second motion vector for the current block from a second list of candidate motion vector predictors based on bilateral matching; and determine a prediction block for the current block using the first motion vector and the second motion vector.
    Type: Application
    Filed: June 12, 2024
    Publication date: January 9, 2025
    Inventors: Zhi Zhang, Chun-Chi Chen, Han Huang, Vadim Seregin, Marta Karczewicz
  • Patent number: 12192469
    Abstract: A method of encoding or decoding video data comprises: for each respective intra prediction mode of a plurality of intra prediction modes in a most-probable mode (MPM) list: generating, based on reference samples for a template region and using the respective intra prediction mode, prediction samples for the template region; and determining a cost for the respective intra prediction mode; determining a first intra prediction mode and a second intra prediction mode in the MPM list having lowest costs; determining a preliminary prediction block for the first intra prediction mode and a preliminary prediction block for the second intra prediction mode; generating a prediction block based on a fusion of the preliminary prediction blocks weighted according to a weight for the first intra prediction mode and a weight for the second intra prediction mode.
    Type: Grant
    Filed: June 1, 2022
    Date of Patent: January 7, 2025
    Assignee: QUALCOMM Incorporated
    Inventors: Keming Cao, Nan Hu, Vadim Seregin, Marta Karczewicz
  • Patent number: 12192311
    Abstract: An example method of encoding a point cloud includes determining that residual values for all components except one component of an attribute of a point in the point cloud are equal to zero; based on the determination that the residual values for all components except the one component of the attribute are equal to zero, determining a value for the one component that is equal to a magnitude of a residual value of the one component of the attribute minus an offset; encoding the value of the one component; and signaling the encoded value in a bitstream.
    Type: Grant
    Filed: June 8, 2021
    Date of Patent: January 7, 2025
    Assignee: QUALCOMM Incorporated
    Inventors: Bappaditya Ray, Geert Van der Auwera, Adarsh Krishnan Ramasubramonian, Louis Joseph Kerofsky, Marta Karczewicz
  • Publication number: 20250008147
    Abstract: A device for decoding video data determines a luma prediction block for a luma block of a current block using an initial motion vector; applies one or more decoder-side motion vector refinement processes to the luma prediction block to determine a refined luma prediction block and refined motion vectors, the one or more motion vector refinement processes comprising a bi-directional optical flow (BDOF) process; for a chroma sample of a chroma block, determines a motion vector for the chroma sample based on refined motion vectors of one or more co-located luma samples of the luma block; determines a chroma prediction sample for a chroma prediction block based on the motion vector for the chroma sample; and determines a decoded version of the current block based on the refined luma prediction block and the chroma prediction block.
    Type: Application
    Filed: June 26, 2024
    Publication date: January 2, 2025
    Inventors: Chun-Chi Chen, Han Huang, Zhi Zhang, Vadim Seregin, Marta Karczewicz
  • Publication number: 20250008134
    Abstract: A device for decoding video data determines a block of a picture; applies a neural network (NN)-based filter process to the block to generate a filtered block, wherein to apply the NN-based filter process, the device performs a first feature extraction on pixel data of the block at a first scale to generate a first set of extracted features for the block; and performs a second feature extraction on the pixel data of the block at a second scale to generate a second set of extracted features for the block, wherein the first scale is different than the second scale; and generates the filtered block based on the first set of extracted features and the second set of extracted features.
    Type: Application
    Filed: June 27, 2024
    Publication date: January 2, 2025
    Inventors: Dmytro Rusanovskyy, Yun Li, Marta Karczewicz
  • Publication number: 20250008113
    Abstract: A video decoder determines that a current block of size width (WCB)×height (HCB) is coded in an affine prediction mode; predicts each subblock of a first plurality of subblocks using an affine motion model to determine an initial prediction block, each subblock having a size of width (WSB)×height (HSB) and WSB being less than WCB and HSB being less than HCB; applies a bi-directional optical flow process to first and second subblocks of a second plurality of subblocks to determine first and second refined prediction subblocks, each subblock having a size of width (WSBIPB)×height (HSBIPB), WSBIPB being less than or equal to WCB and less than or equal to WSB and HSBIPB being less than or equal to HCB and less than or equal to HSB; and determines a refined prediction block based on the first refined subblock and the second refined subblock.
    Type: Application
    Filed: June 26, 2024
    Publication date: January 2, 2025
    Inventors: Zhi Zhang, Han Huang, Jian-Liang Lin, Vadim Seregin, Marta Karczewicz
  • Publication number: 20250008087
    Abstract: An example device includes a memory configured to store video data; and a processing system comprising one or more processors implemented in circuitry and configured to: construct a merge candidate list for a current block of video data, wherein the processing system is configured to add a first merge candidate that was predicted using a first convolutional cross component model (CCCM) to the merge candidate list and add a second merge candidate that was predicted using a second CCCM to the merge candidate list, the first CCCM being different than the second CCCM; decode a merge index value for the current block of video data, the merge index value indicating the first merge candidate; in response to the merge index value indicating the first merge candidate, form a prediction block for the current block using the first CCCM; and decode the current block using the prediction block.
    Type: Application
    Filed: June 27, 2024
    Publication date: January 2, 2025
    Inventors: Han Huang, Vadim Seregin, Chun-Chi Chen, Marta Karczewicz
  • Publication number: 20250008085
    Abstract: A method of encoding or decoding video data includes storing a block vector (BV) for a current block in a history BV list for encoding or decoding a subsequent block, wherein the current block is encoded or decoded in intra temporal motion vector prediction (IntraTMP) mode, and wherein the history BV list includes BVs for blocks that do not neighbor the subsequent block; deriving a candidate list of BVs for the subsequent block based on BVs from the history BV list that includes the BV for the current block; and encoding or decoding the subsequent block based on the candidate list of BVs.
    Type: Application
    Filed: June 26, 2024
    Publication date: January 2, 2025
    Inventors: Po-Han Lin, Jian-Liang Lin, Yao-Jen Chang, Vadim Seregin, Marta Karczewicz
  • Publication number: 20250008082
    Abstract: An example device for decoding video data includes a memory configured to store video data; and a processing system comprising one or more processors implemented in circuitry, the processing system being configured to: determine that vector information for a current block of video data is to be coded using a merge with vector difference mode; determine a search process to be used to determine a vector difference for the vector information; select a merge candidate from a merge candidate list to determine a vector predictor for the vector information; perform the search process to determine the vector difference; apply the vector difference to the vector predictor to form a final vector, and form a prediction block for the current block using the final vector.
    Type: Application
    Filed: May 23, 2024
    Publication date: January 2, 2025
    Inventors: Gleb Verba, Zhi Zhang, Vadim Seregin, Marta Karczewicz
  • Patent number: 12184853
    Abstract: A video coder may determine a quantization parameter (QP) value for a block of video data, determine a residual coding method from a plurality of residual coding methods based on the QP value, wherein the plurality of residual coding methods include transform skip (TS) residual coding and regular residual coding, and code a residual of the block of video data using the determined residual coding method.
    Type: Grant
    Filed: December 2, 2020
    Date of Patent: December 31, 2024
    Assignee: QUALCOMM Incorporated
    Inventors: Alican Nalci, Hongtao Wang, Muhammed Zeyd Coban, Marta Karczewicz