公开(公告)号：US20220036506A1

公开(公告)日：2022-02-03

申请号：US17299557

申请日：2019-12-20

Applicant: ZTE Corporation

Inventor： Dehui KONG ,  Ke XU ,  Xiao ZHANG ,  Bin HAN ,  Zhou HAN ,  Hong WANG ,  Guoning LU ,  Long HUANG ,  Sheng LUO

IPC: G06T3/40 ,  G06T5/50 ,  G06K9/62

Abstract: An image reconstruction method, device and apparatus and non-transitory computer-readable storage medium are disclosed. The method may include: determining norms of convolution kernels of each convolutional layer of a deep neural network model; determining the convolution kernels with norms greater than or equal to a preset threshold in each convolutional layer to obtain a target convolution kernel set of each convolutional layer; processing an input image of each convolutional layer by using the convolution kernels in the target convolution kernel set of each convolutional layer respectively, to obtain a first image processing result; obtaining a second image processing result by performing interpolation on an initial image; and determining a fusion result according to the first image processing result and the second image processing result and reconstructing the initial image according to the fusion result.

公开(公告)号：US20220014745A1

公开(公告)日：2022-01-13

申请号：US17295266

申请日：2019-11-20

Applicant: ZTE Corporation

Inventor： Hong WANG ,  Ke XU ,  Bo HUANG ,  Chengqiang LIU ,  Sheng LUO ,  Guoning LU ,  Degen ZHEN

IPC: H04N19/124 ,  H04N19/60 ,  H04N19/85 ,  H04N19/42

Abstract: Provided are method and device for inverse quantization and inverse transformation and non-transitory computer-readable storage medium. The method includes: performing inverse quantization processing on input data; determining whether secondary inverse transform processing is necessary; performing, in response to determining that the secondary inverse transform processing is necessary, the secondary inverse transform processing on data obtained through the inverse quantization processing, first one-dimensional inverse transform control processing on data obtained through the secondary inverse transform processing, and second one-dimensional inverse transform control processing on data obtained through the first one-dimensional inverse transform control processing; and performing, in response to determining that the secondary inverse transform processing is unnecessary, first one-dimensional inverse transform control processing on the data obtained through the inverse quantization processing, and second one-dimensional inverse transform control processing on the data obtained through the first one-dimensional inverse transform control processing.