The invention discloses a video processing method, and the method comprises the steps that a VR terminal can obtain N picture group sequences of S subareas corresponding to a visual angle of a user after turning from streaming media equipment through a request, wherein the N picture group sequences comprise S picture groups in one-to-one correspondence with the S subareas, and the serial numbers of the S picture groups are the same; and then the VR terminal displays the video contents of the S picture groups in the S subareas in a one-to-one correspondence manner in a panoramic manner. The picture groups with the same serial number in each sub-region are interleaved and transmitted together in an interleaving transmission mode, so the picture groups with the same serial number can be simultaneously displayed in each corresponding sub-region, the problem of difficulty in synchronization is solved, in addition, a group of GOPs with the same serial number can be displayed after being decoded, and the low MTHR is ensured.

Video processing method and device

The invention discloses a JPEG compressed image super-resolution reconstruction method based on a convolutional neural network (CNN). The method mainly includes the following steps of: constructing asuper-resolution reconstruction model based on the CNN for a JPEG compressed image; training the constructed CNN by using a training image; reconstructing a JPEG-compressed low-resolution image by using a trained CNN model. The constructed CNN framework comprises a decompression effect network, a resolution enhancement network and a quality enhancement network, and can be subjected to end-to-end optimization training. The method of the present invention can reduce the compression noise in the JPEG compressed image and improve the resolution of the JPEG compressed image, can be applied to the fields of image and video compression, digital multimedia communication and the like.

JPEG compressed image super-resolution reconstruction method based on convolutional neural network

The invention provides a video super-resolution processing method and device, and the method comprises the steps: obtaining each frame of image of a video, and inputting the image into a convolutionalneural network; Sequentially performing feature extraction, feature dimension reduction, nonlinear mapping and high-dimensional space mapping on the image through the convolutional neural network toobtain super-resolution features; Performing reconstruction according to the features obtained by feature extraction and the super-resolution features to obtain a super-resolution image; And finally,performing coding to form a super-resolution video code stream. The super-resolution processing of the video is realized through the convolutional neural network; Dimension reduction through features,nonlinear mapping and high-dimensional space mapping are carried out; the calculation complexity is reduced, the time complexity is reduced, the network learning difficulty is reduced and the complextexture of the output image is reserved by adopting the jump connection, so that the high reconstruction quality is realized while the real-time performance required by video processing is ensured, and the method has a very wide application prospect in the fields of video real-time transmission and compression, video restoration and the like.

A video super-resolution processing method and device

The invention discloses a super-resolution technology-based picture compression method which comprises the following steps: 1, performing down sampling on a picture to be uploaded by a user of a mobile Internet platform according to a zoom scale, thus obtaining a low-resolution picture, then compressing the low-resolution picture, and uploading the picture; 2, storing and distributing the compressed low-resolution picture through the mobile Internet platform; 3, downloading the compressed low-resolution picture by the user of the mobile Internet platform through different receiving terminals, and performing corresponding decoding on the downloaded compressed low-resolution picture by the receiving terminals, thus obtaining the low-resolution picture; 4, reducing and rebuilding the decoded low-resolution picture into a high-resolution picture for user browsing through a deep convolutional neural network model, wherein the deep convolutional neural network model comprises a bottom layer feature extraction stage, a dense network module-based high layer feature extraction stage, a feature layer fusion stage, a feature layer dimension reduction stage, a deconvolution stage and a rebuilding stage According to the super-resolution technology-based picture compression method disclosed by the invention, the picture can be uploaded, downloaded and browsed more quickly; the storage cost and the bandwidth cost can be saved for at least 50 percent; the original encoding and decoding formats of the picture are not changed; the super-resolution technology-based picture compression method can be seamlessly compatible with all terminals

Super-resolution technology-based picture compression method

The invention is applicable to the field of computer technology and provides a compression recovery method and device for a depth image, a mobile terminal and a storage medium. The method comprises the steps that an input recovery request for a to-be-recovered depth image is received, the to-be-recovered depth image is associated with a corresponding texture image, the texture image and the to-be-recovered depth image are preprocessed, high-frequency information of the texture image and high-frequency information of the to-be-recovered depth image are acquired, the high-frequency information of the texture image, the high-frequency information of the to-be-recovered depth image and the to-be-recovered depth image are input into a Y-branch model, a D-branch model and an M-branch model of a preset depth image recovery model respectively, a feature image obtained after recovery through the Y-branch model and the D-branch model are input into the M-branch model, and the depth image corresponding to the to-be-recovered depth image is recovered through the M-branch model. Therefore, compression recovery quality of the depth image is improved, and accordingly user experience is improved.

Compression recovery method and device for depth image, equipment and storage medium

The invention discloses a video enhancement and transmission method based on deep learning. Downsampling is carried out on a high-definition video at a video source end, thereby obtaining a low-definition video; the low-definition video is compressed in an existing video coding mode; and the compressed low-definition video is transmitted. The size of the video data is greatly reduced through downsampling and video coding, so the video traffic needing to be transmitted is correspondingly reduced, and an effect of reducing the bandwidth cost is achieved. At a user receiving end, a user receives the low-definition video, reconstructs the low-definition video by employing a super-resolution image reconstruction method of the deep learning and restores the low-definition video into a high-resolution video for the user to watch, so the video transmission bandwidth cost is effectively reduced. According to the method, the video transmission bandwidth cost is reduced by at least 50%; the resolution of the live broadcast video is improved; and the watching experience of a user is improved.

Tong Tong

Papers

Video enhancement and transmission method based on deep learning