中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

7 records (took 0.014 seconds)

    ENHANCED ARCHITECTURE FOR DEEP LEARNING-BASED VIDEO PROCESSING
    1.
    发明公开
    ENHANCED ARCHITECTURE FOR DEEP LEARNING-BASED VIDEO PROCESSING 审中-公开

    公开(公告)号：US20240273684A1

    公开(公告)日：2024-08-15

    申请号：US18569528

    申请日：2021-12-10

    Applicant: INTEL CORPORATION

    Inventor： Chen WANG ,  Yi-Jen CHIU ,  Huan DOU ,  Ying ZHANG

    IPC: G06T5/60 ,  G06T5/20

    CPC classification number: G06T5/60 ,  G06T5/20 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/20084

    Abstract: This disclosure describes systems. methods. and devices related to deep learning-based video processing. A system may include a first neural network associated with generating kernel weights for the DL VP. the first neural network using a first hardware device: and a second neural network associated with filtering image pixels for the DLVP. the second neural network using a second hardware device, wherein the first neural network receives image data and generates the kernel weights based on the image data, and wherein the second neural network receives the image data and the kernel weights. and generates filtered image data based on the image data and the kernel weights.

    VARYING CHANNEL WIDTH IN THREE-DIMENSIONAL MEMORY ARRAY
    4.
    发明申请
    VARYING CHANNEL WIDTH IN THREE-DIMENSIONAL MEMORY ARRAY 有权

    公开(公告)号：US20230033086A1

    公开(公告)日：2023-02-02

    申请号：US17791175

    申请日：2020-02-07

    Applicant: Intel Corporation

    Inventor： Chen WANG ,  Dipanjan BASU ,  Richard FASTOW ,  Dimitri KIOUSSIS ,  Yi LI ,  Ebony Lynn MAYS ,  Dimitrios PAVLOPOULOS ,  Junyen TEWG

    IPC: H01L27/11556 ,  H01L27/11582 ,  H01L27/11524 ,  H01L27/1157 ,  G11C8/14

    Abstract: A memory array including a varying width channel is disclosed. The array includes a plurality of WLs, which are above a layer, where the layer can be, for example, a Select Gate Source (SGS) of the memory array, or an isolation layer to isolate a first deck of the array from a second deck of the array. The channel extends through the plurality of word lines and at least partially through the layer. In an example, the channel comprises a first region and a second region. The first region of the channel has a first width that is at least 1 nm different from a second width of the second region of the channel. In an example, the first region extends through the plurality of word lines, and the second region extends through at least a part of the layer underneath the plurality of word lines. In one case, the first width is at least 1 nm less than a second width of the second region of the channel.

    TRAJECTORY PLANNING FOR VIDEO STABILIZATION
    7.
    发明申请
    TRAJECTORY PLANNING FOR VIDEO STABILIZATION 审中-公开

    公开(公告)号：US20170230581A1

    公开(公告)日：2017-08-10

    申请号：US15502990

    申请日：2014-09-19

    Applicant: INTEL CORPORATION

    Inventor： Xiaoxia CAI ,  Chen WANG ,  Lidong XU ,  Yi-Jen CHIU

    IPC: H04N5/232

    CPC classification number: H04N5/23267 ,  H04N5/23254

    Abstract: Trajectory planning is described for video stabilization. In one example, a sequence of video frames is received and low-pass filtered. If a subset of the low-pass filtered frames are determined to be zero-motion frames, then a smooth trajectory is estimated for the for the zero-motion frames. Jitter is removed from the zero-motion frames using the smooth trajectory and frame warping is applied to the zero-motion frames with jitter removed.

Patent Agency Ranking