中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

11 records (took 0.023 seconds)

    METHOD AND APPARATUS FOR ENCODING AND DECODING DIGITAL IMAGES OR VIDEO STREAMS
    1.
    发明申请
    METHOD AND APPARATUS FOR ENCODING AND DECODING DIGITAL IMAGES OR VIDEO STREAMS 审中-公开

    公开(公告)号：WO2018158735A1

    公开(公告)日：2018-09-07

    申请号：PCT/IB2018/051333

    申请日：2018-03-02

    Applicant: SISVEL TECHNOLOGY S.R.L.

    Inventor： GRANGETTO, Marco ,  VERDOJA, Francesco

    IPC: H03M7/30 ,  H04N19/00

    Abstract: A method for encoding digital images or video streams, comprises: - a receiving phase, wherein a portion of an image (f) is received; - a graph weights prediction (GWP) phase, wherein the elements of a weights matrix (W) associated to the graph related to the blocks of the image (predicted blocks) are predicted on the basis of reconstructed, de-quantized and inverse-transformed pixel values of at least one previously coded block (predictor block) of the image (f), the weights matrix (W) being a matrix comprising elements (wij) denoting the level of similarity between a pair of pixels composing said image (f), - a graph transform computation phase, wherein the graph Fourier transform of the blocks of the image (f) is performed, obtaining for said blocks a set of coefficients determined on the basis of the predicted weights; - a coefficients quantization phase, wherein said coefficients are quantized - an output phase wherein a bitstream comprising said transformed and quantized coefficients is transmitted and/or stored.

    METHOD FOR GENERATING AND RECONSTRUCTING A THREE-DIMENSIONAL VIDEO STREAM, BASED ON THE USE OF THE OCCLUSION MAP, AND CORRESPONDING GENERATING AND RECONSTRUCTING DEVICE
    2.
    发明申请
    METHOD FOR GENERATING AND RECONSTRUCTING A THREE-DIMENSIONAL VIDEO STREAM, BASED ON THE USE OF THE OCCLUSION MAP, AND CORRESPONDING GENERATING AND RECONSTRUCTING DEVICE 审中-公开
    用于生成和重新构建三维视频流的方法，基于使用OCCLUSION MAP，以及相应的生成和重新配置设备

    公开(公告)号：WO2013168091A1

    公开(公告)日：2013-11-14

    申请号：PCT/IB2013/053660

    申请日：2013-05-07

    Applicant: SISVEL TECHNOLOGY S.R.L. ,  S.I.SV.EL SOCIETA' ITALIANA PER LO SVILUPPO DELL'ELETTRONICA S.P.A.

    Inventor： GRANGETTO, Marco ,  LUCENTEFORTE, Maurizio

    IPC: H04N7/26 ,  H04N13/00

    CPC classification number: H04N19/597 ,  H04N13/128 ,  H04N13/161 ,  H04N19/547 ,  H04N19/553 ,  H04N2013/0081

    Abstract: A method is described for generating a three-dimensional video stream by starting from a sequence of video images, said sequence comprising a first view (V 0 ), at least one second view (V i ) of a scene, as well as a depth map (D 0 ) of said first view (V o ), or a disparity map of said at least one second view (V 1 ) with respect to the first view (V 0 ), the method comprising, for one image, the following steps: obtaining at least one occlusion image (O 1 ) comprising the occluded pixels of said second view (V 1 ) by starting from said depth map (D 0 ) or from said disparity map; generating a compacted occlusion image (OC 1 ) by spatially repositioning said occluded pixels of said at least one occlusion image (O 1 ), so as to move said pixels closer to one another; said three-dimensional video stream comprising, for one image, said first view (V 0 ), said depth map (D 0 ), or said disparity map, and said at least one compacted occlusion image (OC 1 ).

    Abstract translation: 描述了一种通过从视频图像序列开始生成三维视频流的方法，所述序列包括第一视图（V0），场景的至少一个第二视图（Vi）以及深度图（ 所述第一视图（Vo）的D0）或所述至少一个第二视图（V1）相对于所述第一视图（V0）的视差图，所述方法包括针对一个图像的以下步骤：获得至少一个 通过从所述深度图（D0）开始或从所述视差图开始包括所述第二视图（V1）的遮挡像素的遮挡图像（O1） 通过空间重新定位所述至少一个遮挡图像（O1）的所述遮挡像素来产生压缩遮挡图像（OC1），以使所述像素彼此靠近; 所述三维视频流包括对于一个图像，所述第一视图（V0），所述深度图（D0）或所述视差图以及所述至少一个压缩的遮挡图像（OC1）。

    METHOD AND APPARATUS FOR PRUNING NEURAL NETWORKS
    3.
    发明申请
    METHOD AND APPARATUS FOR PRUNING NEURAL NETWORKS 审中-公开

    公开(公告)号：WO2021074743A1

    公开(公告)日：2021-04-22

    申请号：PCT/IB2020/059379

    申请日：2020-10-06

    Applicant: SISVEL TECHNOLOGY S.R.L. ,  INSTITUT MINES TÉLÉCOM

    Inventor： TARTAGLIONE, Enzo ,  GRANGETTO, Marco ,  ODIERNA, Francesco ,  BRAGAGNOLO, Andrea ,  FIANDROTTI, Attilio

    IPC: G06N3/04 ,  G06N3/08

    Abstract: The present invention relates to a method for pruning a neural network (100) comprising a plurality of neurons (105), said method comprising:- an initialization phase, wherein input information is fetched comprising at least parameters ({wni,bni}) related to said neural network (100) and a dataset (D) representative of a task that said neural network (100) has to deal with, wherein said parameters ({wni,bni}) comprising a weights vector (wni) and/or a bias (bni) related to at least one neuron (105) of said plurality of neurons;- a regularization phase, wherein said neural network (100) is trained according to a training algorithm by using said dataset (D);- a thresholding phase, wherein an element (wnij) of said weights vector (wni) is put at zero when its absolute value is below a given threshold (T),said method wherein, during said regularization phase, said parameters ({wni,bni}) evolve according to a regularized update rule based on a neural sensitivity measure (S) to drive towards zero parameters related to at least one less sensitive neuron (108) of said neural network (100), wherein said neural sensitivity measure (S) is based on a pre-activation signal of at least one neuron (105) of said plurality of neurons.

    METHODS AND APPARATUSES FOR ENCODING AND DECODING SUPERPIXEL BORDERS
    4.
    发明申请
    METHODS AND APPARATUSES FOR ENCODING AND DECODING SUPERPIXEL BORDERS 审中-公开

    公开(公告)号：WO2018158738A1

    公开(公告)日：2018-09-07

    申请号：PCT/IB2018/051339

    申请日：2018-03-02

    Applicant: SISVEL TECHNOLOGY S.R.L.

    Inventor： GRANGETTO, Marco ,  VERDOJA, Francesco

    IPC: H03M7/30 ,  G06T9/20 ,  G06K9/48

    Abstract: The present invention relates to a method for encoding the borders of pixel regions of an image, wherein the borders contain a sequence of vertices subdividing the image into regions of pixels (superpixels), by generating a sequence of symbols from an alphabet comprising the step of: - defining for each superpixel a first vertex for coding the borders of the superpixel according to a criterion common to all superpixels; - defining for each superpixel the same coding order of the border vertices, either clockwise or counter-clockwise; - defining the order for coding the superpixels on the base of a common rule depending on the relative positions of the first vertices; - defining a set of vertices as a known border ( Γ ), wherein the following steps are performed for selecting a symbol of the alphabet, for encoding the borders of the superpixels: a) determining the first vertex of the next superpixel border individuated by the common criterion; b) determining the next vertex to be encoded on the basis of the coding direction; c) selecting a first symbol ("0") for encoding the next vertex if the next vertex of a border pertains to the known border ( Γ ), d) selecting a symbol ("1"; "2") different from the first symbol ("0") if the next vertex is not in the known border (( Γ ); e) repeating steps b), c), d) and e) until all vertices of the superpixel border have been encoded; f) adding each vertex of the superpixel border that was not in the known border ( Γ ) to the set (( Γ ); g) determining the next superpixel whose border is to be encoded according to the common rule, if any; i) repeating steps a)-g) until the borders of all the superpixels of the image have being added to the known border ( Γ ).

    APPARATUS AND METHOD FOR IMAGE SEGMENTATION
    5.
    发明公开
    APPARATUS AND METHOD FOR IMAGE SEGMENTATION 审中-公开
    BILDSEGMENTIERUNGSVORRICHTUNG以外地区

    公开(公告)号：EP3018626A1

    公开(公告)日：2016-05-11

    申请号：EP15191591.5

    申请日：2015-10-27

    Applicant: Sisvel Technology S.r.l.

    Inventor： GRANGETTO, Marco ,  FARID, Muhammad Shahid

    IPC: G06T7/00

    CPC classification number: G06T7/162 ,  G06T7/143 ,  G06T7/155 ,  G06T7/187 ,  G06T7/194 ,  G06T2207/20036 ,  G06T2207/20072 ,  G06T2207/20076 ,  G06T2207/20101

    Abstract: The present invention relates to an apparatus and a method for image segmentation, which comprises a pre-segmentation phase (P1), wherein a set of points of said image is selected, a morphological expansion phase (P3), wherein the set of points selected during the pre-segmentation phase (P1) is expanded by taking into account the spatial distribution of the points of the selected set in the image, a graph generation phase (SP41), wherein a graph (GR) is generated on the basis of the image and the selected set of points, a graph partitioning phase (SP42), wherein a set of nodes of the graph (GR) is selected on the basis of the characteristics of said graph (GR) and the points corresponding to the selected nodes are added to the selected set of points.

    Abstract translation: 本发明涉及一种用于图像分割的装置和方法，其包括预分割阶段（P1），其中选择所述图像的一组点，形态扩展阶段（P3），其中所选择的点集合 通过考虑图像中所选集合的点的空间分布来扩展预分割阶段（P1）期间的图生成阶段（SP41），其中基于 图像和所选择的点集合，图形分割阶段（SP42），其中基于所述图形（GR）的特征和对应于所选择的节点的点来选择图形（GR）的一组节点 添加到所选择的一组点。

    APPARATUS AND METHOD FOR IMAGE SEGMENTATION
    8.
    发明公开
    APPARATUS AND METHOD FOR IMAGE SEGMENTATION 审中-公开
    用于图像分割的装置和方法

    公开(公告)号：EP3018626A9

    公开(公告)日：2016-06-29

    申请号：EP15191591.5

    申请日：2015-10-27

    Applicant: Sisvel Technology S.r.l.

    Inventor： GRANGETTO, Marco ,  FARID, Muhammad Shahid

    IPC: G06T7/00

    CPC classification number: G06T7/162 ,  G06T7/143 ,  G06T7/155 ,  G06T7/187 ,  G06T7/194 ,  G06T2207/20036 ,  G06T2207/20072 ,  G06T2207/20076 ,  G06T2207/20101

    Abstract: The present invention relates to an apparatus and a method for image segmentation, which comprises a pre-segmentation phase (P1), wherein a set of points of said image is selected, a morphological expansion phase (P3), wherein the set of points selected during the pre-segmentation phase (P1) is expanded by taking into account the spatial distribution of the points of the selected set in the image, a graph generation phase (SP41), wherein a graph (GR) is generated on the basis of the image and the selected set of points, a graph partitioning phase (SP42), wherein a set of nodes of the graph (GR) is selected on the basis of the characteristics of said graph (GR) and the points corresponding to the selected nodes are added to the selected set of points.

    Abstract translation: 本发明涉及用于图像分割的装置和方法，其包括预分割阶段（P1），其中选择所述图像的一组点，形态学扩展阶段（P3），其中选择的点集合 在预分割阶段（P1）期间，通过考虑图像中所选择的组的点的空间分布来扩展（P1），图生成阶段（SP41），其中图（GR）基于 图像和所选择的一组点，图分割阶段（SP42），其中基于所述图（GR）的特征选择图（GR）的一组节点并且与所选节点对应的点是 添加到选定的一组点。

Patent Agency Ranking