公开(公告)号：US11930215B2

公开(公告)日：2024-03-12

申请号：US17448658

申请日：2021-09-23

Applicant: QUALCOMM Incorporated

Inventor： Hongtao Wang ,  Venkata Meher Satchit Anand Kotra ,  Jianle Chen ,  Marta Karczewicz ,  Dana Kianfar ,  Auke Joris Wiggers

IPC: H04N19/70 ,  H04N19/172 ,  H04N19/192 ,  H04N19/44 ,  H04N19/80

CPC classification number: H04N19/70 ,  H04N19/172 ,  H04N19/192 ,  H04N19/44 ,  H04N19/80

Abstract: An example device for filtering decoded video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: decode a picture of video data; code a value for a syntax element representing a neural network model to be used to filter a portion of the decoded picture, the value representing an index into a set of pre-defined neural network models, the index corresponding to the neural network model in the set of pre-defined neural network models; and filter the portion of the decoded picture using the neural network model corresponding to the index.

公开(公告)号：US20240305785A1

公开(公告)日：2024-09-12

申请号：US18457079

申请日：2023-08-28

Applicant: QUALCOMM Incorporated

Inventor： Ties Jehan Van Rozendaal ,  Hoang Cong Minh Le ,  Tushar Singhal ,  Amir Said ,  Krishna Buska ,  Guillaume Konrad Sautiere ,  Anjuman Raha ,  Auke Joris Wiggers ,  Frank Steven Mayer ,  Liang Zhang ,  Abhijit Khobare ,  Muralidhar Reddy Akula

IPC: H04N19/137 ,  H04N19/159 ,  H04N19/176 ,  H04N19/192

CPC classification number: H04N19/137 ,  H04N19/159 ,  H04N19/176 ,  H04N19/192

Abstract: An example computing device may include memory and one or more processors. The one or more processors may be configured to parallel entropy decode encoded video data from a received bitstream to generate entropy decoded data. The one or more processors may be configured to predict a motion vector based on the entropy decoded data. The one or more processors may be configured to decode a motion vector residual from the entropy decoded data. The one or more processors may be configured to add the motion vector residual and motion vector. The one or more processors may be configured to warp previous reconstructed video data with an overlapped block-based warp function using the motion vector to generate predicted current video data. The one or more processors may be configured to sum the predicted current video data with a residual block to generate current reconstructed video data.

公开(公告)号：US12058348B2

公开(公告)日：2024-08-06

申请号：US17070589

申请日：2020-10-14

Applicant: QUALCOMM Incorporated

Inventor： Dana Kianfar ,  Auke Joris Wiggers ,  Amir Said ,  Taco Sebastiaan Cohen ,  Reza Pourreza Shahri

IPC: H04N19/124 ,  G05B13/00 ,  G06F30/27 ,  G06N3/02 ,  G06N3/047 ,  H04N19/13 ,  H04N19/176 ,  H04N19/18 ,  H04N19/61

CPC classification number: H04N19/176 ,  G05B13/00 ,  G06F30/27 ,  G06N3/02 ,  G06N3/047 ,  H04N19/124 ,  H04N19/13 ,  H04N19/18 ,  H04N19/61 ,  H04Q2213/13343 ,  H04Q2213/343 ,  Y10S706/00

Abstract: A video encoder determines scaled transform coefficients, wherein determining the scaled transform coefficients comprises scaling transform coefficients of a block of the video data according to a given quantization step. The video encoder determines scalar quantized coefficients, wherein determining the scalar quantized coefficients comprises applying scalar quantization to the scaled transform coefficients of the block. Additionally, the video encoder applies a neural network that determines a respective set of probabilities for each respective transform coefficient of the block. The respective set of probabilities for the respective transform coefficient includes a respective probability value for each possible adjustment value in a plurality of possible adjustment values. Inputs to the neural network include the scaled transform coefficients and the scalar quantized coefficients. The video encoder determines, based on the set of probabilities for a particular transform coefficient of the block, a quantization level for the particular transform coefficient.

公开(公告)号：US12177473B2

公开(公告)日：2024-12-24

申请号：US17862217

申请日：2022-07-11

Applicant: QUALCOMM Incorporated

Inventor： Reza Pourreza ,  Hoang Cong Minh Le ,  Auke Joris Wiggers

IPC: H04N19/52 ,  H04N19/105 ,  H04N19/137 ,  H04N19/172

Abstract: Systems and techniques are provided for coding video data based on an optical flow correction and a residual correction. For example, a decoding device can obtain a frame of encoded video data associated with an input frame, the frame of encoded video data including an optical flow correction and a residual correction. A predicted optical flow can be generated based on one or more reference frames and a reference optical flow. A corrected prediction frame can be generated based on the predicted optical flow and the optical flow correction. A predicted residual can be generated based on at least the corrected prediction frame and a first reference frame included in the one or more reference frames. The decoding device can generate a reconstructed input frame based on the corrected prediction frame, the predicted residual, and the residual correction.

公开(公告)号：US20220103864A1

公开(公告)日：2022-03-31

申请号：US17448658

申请日：2021-09-23

Applicant: QUALCOMM Incorporated

Inventor： Hongtao Wang ,  Venkata Meher Satchit Anand Kotra ,  Jianle Chen ,  Marta Karczewicz ,  Dana Kianfar ,  Auke Joris Wiggers

IPC: H04N19/70 ,  H04N19/44 ,  H04N19/172 ,  H04N19/192 ,  H04N19/80

Abstract: An example device for filtering decoded video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: decode a picture of video data; code a value for a syntax element representing a neural network model to be used to filter a portion of the decoded picture, the value representing an index into a set of pre-defined neural network models, the index corresponding to the neural network model in the set of pre-defined neural network models; and filter the portion of the decoded picture using the neural network model corresponding to the index.

公开(公告)号：US20210329267A1

公开(公告)日：2021-10-21

申请号：US17070589

申请日：2020-10-14

Applicant: QUALCOMM Incorporated

Inventor： Dana Kianfar ,  Auke Joris Wiggers ,  Amir Said ,  Taco Sebastiaan Cohen ,  Reza Pourreza Shahri

IPC: H04N19/176 ,  H04N19/61 ,  H04N19/13 ,  H04N19/18 ,  H04N19/124

Abstract: A video encoder determines scaled transform coefficients, wherein determining the scaled transform coefficients comprises scaling transform coefficients of a block of the video data according to a given quantization step. The video encoder determines scalar quantized coefficients, wherein determining the scalar quantized coefficients comprises applying scalar quantization to the scaled transform coefficients of the block. Additionally, the video encoder applies a neural network that determines a respective set of probabilities for each respective transform coefficient of the block. The respective set of probabilities for the respective transform coefficient includes a respective probability value for each possible adjustment value in a plurality of possible adjustment values. Inputs to the neural network include the scaled transform coefficients and the scalar quantized coefficients. The video encoder determines, based on the set of probabilities for a particular transform coefficient of the block, a quantization level for the particular transform coefficient.

公开(公告)号：US20240244265A1

公开(公告)日：2024-07-18

申请号：US18433946

申请日：2024-02-06

Applicant: QUALCOMM Incorporated

Inventor： Hongtao Wang ,  Venkata Meher Satchit Anand Kotra ,  Jianle Chen ,  Marta Karczewicz ,  Dana Kianfar ,  Auke Joris Wiggers

IPC: H04N19/70 ,  H04N19/172 ,  H04N19/192 ,  H04N19/44 ,  H04N19/80

CPC classification number: H04N19/70 ,  H04N19/172 ,  H04N19/192 ,  H04N19/44 ,  H04N19/80

Abstract: An example device for filtering decoded video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: decode a picture of video data; code a value for a syntax element representing a neural network model to be used to filter a portion of the decoded picture, the value representing an index into a set of pre-defined neural network models, the index corresponding to the neural network model in the set of pre-defined neural network models; and filter the portion of the decoded picture using the neural network model corresponding to the index.

公开(公告)号：US11899411B2

公开(公告)日：2024-02-13

申请号：US18049263

申请日：2022-10-24

Applicant: Qualcomm Incorporated

Inventor： Mohammad Naghshvar ,  Ahmed Kamel Sadek ,  Auke Joris Wiggers

IPC: G05B13/02 ,  G05D1/02 ,  G05D1/00 ,  G06N3/045 ,  G06V10/764 ,  G06V10/80 ,  G06V10/00 ,  G06V20/05

CPC classification number: G05B13/027 ,  G05D1/0088 ,  G05D1/0221 ,  G05D1/0231 ,  G05D1/0257 ,  G06N3/045 ,  G06V10/00 ,  G06V10/764 ,  G06V10/811 ,  G06V20/05

Abstract: A method includes determining a current state of an environment of an autonomous agent, such as a vehicle. The method also includes determining, via a first neural network, a set of actions based on the current state. The method further includes determining whether further analysis of the set of actions is desired. The method selects an action from the set of actions using a model-based solution based on a reward and a risk of the action when further analysis is desired. The method also includes selecting the action from the set of actions according to a metric when further analysis is not desired. The method controls the autonomous agent to perform the selected action.

公开(公告)号：US11480972B2

公开(公告)日：2022-10-25

申请号：US16683129

申请日：2019-11-13

Applicant: QUALCOMM Incorporated

Inventor： Mohammad Naghshvar ,  Ahmed Kamel Sadek ,  Auke Joris Wiggers

IPC: G05D1/02 ,  G05B13/02 ,  G05D1/00 ,  G06N3/04

Abstract: A method includes determining a current state of an environment of an autonomous agent, such as a vehicle. The method also includes determining, via a first neural network, a set of actions based on the current state. The method further includes determining whether further analysis of the set of actions is desired. The method selects an action from the set of actions using a model-based solution based on a reward and a risk of the action when further analysis is desired. The method also includes selecting the action from the set of actions according to a metric when further analysis is not desired. The method controls the autonomous agent to perform the selected action.