3D prediction method for video coding

    公开(公告)号:US12063389B2

    公开(公告)日:2024-08-13

    申请号:US17984997

    申请日:2022-11-10

    申请人: Rovi Guides, Inc.

    摘要: Systems and methods are provided for using a Multiple Depth Plane (MDP) prediction in predictive coding. The system detects a camera viewpoint change between a current frame and a previous frame, decomposes a reconstructed depth map of the previous frame to a plurality of depth planes, adjusts the plurality of depth planes from a previous camera viewpoint to correspond with a current camera viewpoint, generates an MDP prediction by summing pixel values of the adjusted plurality of depth planes along a plurality of optical axes from the current camera viewpoint, determines an MDP prediction error between the MDP prediction and a depth map of the current frame, quantizes and codes the MDP prediction error, and transmits, to a receiver over a communication network, the camera viewpoint change and the coded quantized MDP prediction error for reconstruction of a depth map of the current frame.

    VIDEO CODEC AWARE RADIO ACCESS NETWORK CONFIGURATION AND UNEQUAL ERROR PROTECTION CODING

    公开(公告)号:US20240187650A1

    公开(公告)日:2024-06-06

    申请号:US18396580

    申请日:2023-12-26

    IPC分类号: H04N19/67 H04L1/00 H04N19/89

    摘要: Various aspects of the present disclosure relate to video codec aware RAN configuration and unequal error protection coding. An apparatus includes a memory and a processor coupled to the memory that is configured to cause the apparatus to detect a video coded traffic stream and a video codec profile for encoding the video coded traffic stream, determine an awareness of PDU sets of the video coded traffic stream, align, based on at least the awareness of PDU sets, each PDU set of the video coded traffic stream to PHY transport elements and channel coding element partitions for a video coded traffic aware PHY transport, determine a channel coding rate allocation of the channel coding element partitions based on the video coded traffic aware PHY transport, and apply a FEC coding to the channel coding element partitions based in part on the channel coding rate allocation.

    SYSTEM AND METHOD FOR CORRECTING NETWORK LOSS OF DATA

    公开(公告)号:US20230188755A1

    公开(公告)日:2023-06-15

    申请号:US17887620

    申请日:2022-08-15

    申请人: Agora Lab, Inc.

    摘要: A reference-order AL-FEC system for recovering network video data packet loss during real-time video communication includes a packetizer, a reference-order AL-FEC encoder, a reference-order AL-FEC decoder and a depacketizer. The packetizer constructs source symbols from source packets of a current frame. The encoder generates a repair symbol from the source symbols of the current frame and other reference frames based on the reference-order, not time-order, between the frames within an encoding window. The encoder also generates a repair packet based on the repair symbol. The decoder recovers a lost source symbol based on the source symbols of the frames of the encoding window and the repair symbol by decoding the repair packet. The decoding is achieved by solving a linear system of the repair symbol.