Patents by Inventor Sehoon Yea

Sehoon Yea 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).

  • Patent number: 11682144
    Abstract: A method and device for coding attribute information of a point cloud may include obtaining the point cloud including a set of points in a three-dimensional space. The attribute information of the point cloud is coded using a YCoCg-R transform.
    Type: Grant
    Filed: September 24, 2020
    Date of Patent: June 20, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu
  • Patent number: 11683513
    Abstract: Systems and methods for encoding a video stream are provided. A method includes signaling partitioning information in a coded bitstream that is based on a point cloud. The coded bitstream may be a coded video stream that includes a frame of a plurality of two-dimensional (2D) pictures that are layers of the frame, each of the plurality of 2D pictures having a respective attribute of a same three-dimensional (3D) representation; frame partition information that indicates the frame is partitioned into a plurality of sub-frames, each of the plurality of sub-frames being a respective combination of a sub-region of each picture of the plurality of 2D pictures; and 3D bounding box information that specifies a 3D position corresponding to a sub-frame of the plurality of sub-frames.
    Type: Grant
    Filed: September 3, 2021
    Date of Patent: June 20, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Byeongdoo Choi, Arash Vosoughi, Sehoon Yea, Shan Liu
  • Patent number: 11671577
    Abstract: Aspects of the disclosure provide methods and apparatuses for point cloud compression and decompression. In some examples, an apparatus for point cloud compression/decompression includes processing circuitry. In some embodiments, the processing circuitry decodes prediction information of a point cloud from a coded bitstream and reconstructs a geometry reconstructed cloud according to a geometry image of the point cloud that is decoded from the coded bitstream. Further, the processing circuitry applies a filter to at least a geometry sample inside a patch of the geometry reconstructed cloud in addition to boundary samples of the patch to generate a smoothed geometry reconstructed cloud, and reconstructs points of the point cloud based on the smoothed geometry reconstructed cloud.
    Type: Grant
    Filed: October 13, 2021
    Date of Patent: June 6, 2023
    Assignee: Tencent America LLC
    Inventors: Arash Vosoughi, Sehoon Yea, Shan Liu
  • Patent number: 11671576
    Abstract: A method of interframe point cloud attribute decoding of video data is performed by at least one processor and includes: obtaining a first reconstructed residual; obtaining a quantization index of a second reconstructed residual; obtaining the second reconstructed residual, based on the obtained quantization index of the second reconstructed residual and the obtained first reconstructed residual; and obtaining a color attribute of a point of a point cloud by decoding the obtained second reconstructed residual or decoding a transform residual based on the obtained second reconstructed residual. The first reconstructed residual and the second reconstructed residual are each a reconstructed residual of a respective channel of the color attribute.
    Type: Grant
    Filed: October 7, 2021
    Date of Patent: June 6, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Arash Vosoughi, Shan Liu
  • Patent number: 11657542
    Abstract: A method, apparatus, and computer-readable medium for point cloud coefficient coding are provided. Transform coefficients associated with point cloud data are decomposed into set-index values and symbol-index values, the symbol index-value specifying location of the transform coefficient within a set. The decomposed transform coefficients are partitioned into one or more sets based on the set-index values and the symbol-index values. The set-index values of the partitioned transform coefficients are entropy-coded, and the symbol-index values of the partitioned transform coefficients are bypass-coded. The point cloud data is compressed based on the entropy-coded symbol-index values and the bypass-coded set-index values.
    Type: Grant
    Filed: November 29, 2022
    Date of Patent: May 23, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu
  • Publication number: 20230156224
    Abstract: A Method of decoding an encoded video bitstream using at least one processor includes: obtaining an encoded video bitstream, the encoded video bitstream including encoded color components; entropy parsing the encoded color components; dequantizing the color components and obtaining transform coefficients of the color components; applying a joint components secondary transform (JCST) on the transform coefficients of the color components, thereby generating JCST outputs; performing a backward transform on the JCST outputs, thereby obtaining residual components of the color components; and decoding the encoded video bitstream based on the residual components of the color components.
    Type: Application
    Filed: January 13, 2023
    Publication date: May 18, 2023
    Applicant: TENCENT AMERICA LLC
    Inventors: Xin ZHAO, Sehoon YEA, Shan LIU
  • Patent number: 11651521
    Abstract: A method, apparatus, and computer-readable medium for point cloud coefficient coding are provided. The method may include receiving compressed point cloud data based on set-index values and symbol-index values; and entropy-decoding the set-index values based on the compressed point cloud data. The symbol-index values may be bypass-decoded based on the compressed point cloud data, and the set-index values and symbol-index values may be combined into transform coefficients associated with cloud point data.
    Type: Grant
    Filed: May 31, 2022
    Date of Patent: May 16, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu
  • Patent number: 11647228
    Abstract: The present invention provides a method for decoding a video signal including extracting an inter prediction mode of a current block from the video signal; generating two non-square prediction blocks using two motion vectors according to the inter prediction mode; performing a non-square transform on the two non-square prediction blocks to obtain two non-square prediction blocks on a frequency domain; updating the two non-square prediction blocks on the frequency domain using a correlation coefficient or a scaling coefficient; and generating a reconstruction block based on the updated non-square prediction blocks on the frequency domain and a residual block.
    Type: Grant
    Filed: July 15, 2016
    Date of Patent: May 9, 2023
    Assignee: LG Electronics Inc.
    Inventors: Jin Heo, Sehoon Yea, Bumshik Lee
  • Patent number: 11616983
    Abstract: A Method of decoding an encoded video bitstream using at least one processor includes: obtaining an encoded video bitstream, the encoded video bitstream including encoded color components; entropy parsing the encoded color components; dequantizing the color components and obtaining transform coefficients of the color components; applying a joint components secondary transform (JCST) on the transform coefficients of the color components, thereby generating JCST outputs; performing a backward transform on the JCST outputs, thereby obtaining residual components of the color components; and decoding the encoded video bitstream based on the residual components of the color components.
    Type: Grant
    Filed: October 16, 2020
    Date of Patent: March 28, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Xin Zhao, Sehoon Yea, Shan Liu
  • Publication number: 20230090878
    Abstract: A method, apparatus, and computer-readable medium for point cloud coefficient coding are provided. Transform coefficients associated with point cloud data are decomposed into set-index values and symbol-index values, the symbol index-value specifying location of the transform coefficient within a set. The decomposed transform coefficients are partitioned into one or more sets based on the set-index values and the symbol-index values. The set-index values of the partitioned transform coefficients are entropy-coded, and the symbol-index values of the partitioned transform coefficients are bypass-coded. The point cloud data is compressed based on the entropy-coded symbol-index values and the bypass-coded set-index values.
    Type: Application
    Filed: November 29, 2022
    Publication date: March 23, 2023
    Applicant: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu
  • Publication number: 20230085672
    Abstract: A method of video decoding includes determining directionality information of a restoration filter unit included in a video frame based on directionality of each of one or more groups of pixels in the restoration filter unit determined by at least one of a previously performed CDEF process or a previously performed intra prediction coding process. The method also includes selecting, based on the directionality information of the restoration filter unit, a set of filter parameters from among a plurality of sets of filter parameters of a restoration filter process, and, after the at least one of the previously performed CDEF process or the previously performed intra prediction coding process, performing one of a Wiener filter process or a self-guided projection filter process as the restoration filter process on the restoration filter unit based on the selected set of filter parameters.
    Type: Application
    Filed: November 28, 2022
    Publication date: March 23, 2023
    Applicant: TENCENT AMERICA LLC
    Inventors: Sehoon YEA, Yixin DU, Xin ZHAO, Shan LIU
  • Patent number: 11611770
    Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for video encoding/decoding. An apparatus includes processing circuitry that determines directionality information of a restoration filter unit included in a video frame based on at least one of a constrained directional enhancement filter (CDEF) process or an intra prediction mode. The processing circuitry determines one of a plurality of sets of filter parameters of a restoration filter process based on the directionality information of the restoration filter unit. The processing circuitry performs the restoration filter process on the restoration filter unit based on the one of the plurality of sets of filter parameters. The processing circuitry reconstructs the video frame based on the filtered restoration filter unit.
    Type: Grant
    Filed: June 29, 2021
    Date of Patent: March 21, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Yixin Du, Xin Zhao, Shan Liu
  • Patent number: 11606561
    Abstract: A method of adaptive point cloud attribute coding is performed by at least one processor and includes determining a centroid of k candidate points of a point cloud, k being a predetermined sampling rate of all points of the point cloud, and selecting one of the k candidate points that is closest to the centroid. The method further includes determining a first levels-of-detail (LoD) layer comprising the one of the k candidate points, and a second LoD layer without the one of the k candidate points, and performing coding of attributes of the all points, based on the first LoD layer and the second LoD layer.
    Type: Grant
    Filed: August 27, 2021
    Date of Patent: March 14, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Birendra Kathariya, Sehoon Yea, Arash Vosoughi, Shan Liu
  • Patent number: 11587263
    Abstract: A method and apparatus for decoding a video stream encoded using video point cloud coding, the decoding including obtaining a geometry-reconstructed point cloud based on one or more patches; identifying a first boundary of a patch including a plurality of first boundary points; identifying a second boundary including a plurality of second boundary points inside the first boundary; performing smoothing on the first boundary points and the second boundary points; obtaining a smoothed geometry-reconstructed point cloud based on the smoothed first boundary points and the smoothed second boundary points; and reconstructing a dynamic point cloud using the smoothed geometry-reconstructed point cloud.
    Type: Grant
    Filed: August 16, 2021
    Date of Patent: February 21, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Arash Vosoughi, Sehoon Yea, Shan Liu
  • Patent number: 11580670
    Abstract: A method of interframe point cloud attribute coding is performed by at least one processor and includes coding a first color attribute of a point of a point cloud to obtain a first reconstructed residual, coding a second color attribute of the point to obtain a second reconstructed residual, and determining a quantization index of the second reconstructed residual, based on the first reconstructed residual and the second reconstructed residual. The method further includes updating the second reconstructed residual, based on the quantization index and the first reconstructed residual.
    Type: Grant
    Filed: December 9, 2021
    Date of Patent: February 14, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Arash Vosoughi, Shan Liu
  • Patent number: 11582484
    Abstract: There is included a method and apparatus comprising computer code configured to cause a processor or processors to perform obtaining a leaf node of geometry based point cloud compression (G-PCC) data, splitting the leaf node into a plurality of cuboids, deriving separate triangle soups for each of the cuboids, and coding a plurality of flags respectively for each of the edges of each of the cuboids, where the plurality of flags indicate whether vertices of the separate triangle soups are present on ones of the edges.
    Type: Grant
    Filed: September 9, 2021
    Date of Patent: February 14, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Arash Vosoughi, Sehoon Yea, Stephan Wenger, Shan Liu
  • Patent number: 11568571
    Abstract: A method of point cloud attribute coding includes obtaining an attribute signal corresponding to a point cloud; determining whether lossless lifting is enabled; based on determining that lossless lifting is enabled, modifying at least one from among a plurality of quantization weight and a plurality of lifting coefficients; decomposing the attribute signal into a plurality of detail signals and a plurality of approximation signals based on the modified at least one from among the plurality of quantization weights and the plurality of lifting coefficients; generating a bitstream representing the point cloud based on the plurality of detail signals and the plurality of approximation signals; and transmitting the bitstream.
    Type: Grant
    Filed: September 24, 2020
    Date of Patent: January 31, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu
  • Publication number: 20230015039
    Abstract: A method of point cloud geometry decoding including receiving, from a coded bitstream for a point cloud that includes a set of points in a three-dimensional (3D) space, first signaling information being indicative of partition information of the point cloud, and receiving second signaling information that indicates whether the 3D space is an asymmetric cuboid. The method further includes determining dimensions of the 3D space that are signaled along x, y, and z directions based on the second signaling information indicating that the 3D space is the asymmetric cuboid. The method further includes, in response to the first signaling information indicating that the 3D space is the asymmetric cuboid, partitioning the 3D space based on the determined dimensions of the 3D space and based on the partition information indicated by the first signaling information, and reconstructing, by the processing circuitry, the point cloud based on the partitioned 3D space.
    Type: Application
    Filed: September 16, 2022
    Publication date: January 19, 2023
    Applicant: Tencent America LLC
    Inventors: Sehoon YEA, Wen GAO, Xiang ZHANG, Shan LIU
  • Patent number: 11551334
    Abstract: A method, computer system, and computer-readable medium are provided for point cloud attribute coding by at least one processor. Data associated with a point cloud is received. The received data is transformed through a lifting decomposition based on enabling a scalable coding of attributes associated with the lifting decomposition. The point cloud is reconstructed based on the transformed data.
    Type: Grant
    Filed: December 29, 2020
    Date of Patent: January 10, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu, Wen Gao
  • Patent number: 11544819
    Abstract: A method, apparatus, and computer-readable medium for point cloud coefficient coding are provided. Transform coefficients associated with point cloud data are decomposed into set-index values and symbol-index values, the symbol index-value specifying location of the transform coefficient within a set. The decomposed transform coefficients are partitioned into one or more sets based on the set-index values and the symbol-index values. The set-index values of the partitioned transform coefficients are entropy-coded, and the symbol-index values of the partitioned transform coefficients are bypass-coded. The point cloud data is compressed based on the entropy-coded symbol-index values and the bypass-coded set-index values.
    Type: Grant
    Filed: January 27, 2022
    Date of Patent: January 3, 2023
    Assignee: TENCENT AMERICA LLC
    Inventors: Sehoon Yea, Stephan Wenger, Shan Liu