Hangzhou Hikvision Digital Technology Co., Ltd.

Xiaoyang Wu

Shiliang Pu

Linjie Shen

Hui Su

Lulu Zhou

Li Wang

Hai Yu

Di Xie

Jianhua Chen

Embodiments of the present application provide a method, apparatus, and electronic device for eliminating distortion of a distorted image. Side information components of a distorted image are generated. The distorted image is resulted from image processing on an original image. Side information components represent distortion features of the distorted image with respect to original image. Distorted image color components of the distorted image and the side information components are input into a pre-established convolutional neural network model for convolution filtering to obtain distortion-eliminated image color components. The convolutional neural network model is obtained through training based on a preset training set. The training set comprises original sample images, distorted image color components of multiple distorted images corresponding to each of the original sample images, and side information components of each of the distorted images. Thereby, a CNN-based distortion elimination process different from related art is provided.

Image distortion correction method and device and electronic device

A bit rate control method device obtains a bit rate control mode and coding control parameters of a current short period in a long period, determines a frame-level coding parameter of the current short period, and transmits the frame-level coding parameter to an encoder. Second bit stream information is obtained and statistical analysis is performed which includes complexity information of the current video frame, the number of the encoded bits in the current short period, and coding quality information of the current short period, so as to encode a video frame of the next, adjacent short period by calculating a frame-level coding parameter of the next, adjacent short period. By the present solution, the storage space is allocated reasonably under the condition that the image satisfies a certain quality.

Bit rate control method and device

When performing an inter predictive encoding on any image frame x is required, a reference image frame is determined for image x; image x is a P-frame image or a B-frame image; for each reference image frame, following processes are respectively performed: whether resolution of a reference image frame is the same as resolution of image x is determined; if resolution of the reference image frame is different from resolution of image x, resolution of reference image frame is adjusted to be the same as resolution of image x; if resolution of each reference image frame is the same as resolution of image x, inter predictive encoding is performed on image x based on each reference image frame.

Inter prediction method and device

Systems, devices and methods are provided for video encoding and decoding. An example method includes: analyzing one or more input images; obtaining a background image from the one or more input images; encoding the background image to obtain a background frame; decoding the background frame to obtain a reconstructed image of the background frame; obtaining a first image from the one or more input images; encoding at least partially the first image to obtain a refreshment frame based at least in part on a difference between the first image and the reconstructed image of the background frame; obtaining a second image from the one or more input images; encoding at least partially the second image to obtain a common frame; and generating a video bit stream based at least in part on the background frame, the refreshment frame and the common frame.

Growing community of inventors

Hangzhou, China

Xiaoyang Wu