A Method of Improving the Spatial Resolution of Video
A spatial resolution and low resolution technology, applied in the field of video editing, can solve the problems of increasing the computational complexity of the super-resolution process, speeding up the video super-resolution, etc., and achieve the goals of simplifying the video super-resolution process, efficient calculation speed, and enhancing video quality Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0038] figure 1 For one frame of a low-resolution video sequence (the nth frame I n ) super-resolution process, the specific steps are as follows:
[0039] (1) First, the current frame I n Two frames of images before and after it (I n-2 , I n-1 , I n+1 , I n+2 ) is aggregated as lr (lr={I n-2 , I n-1 , I n , I n+1 , I n+2}). Then input lr into the motion compensation network, and output 8 pairs of one-dimensional convolution kernels (v 1 , h 1 ), (v 2 , h 2 ),..., (v 8 , h 8 ). A total of 32 pairs of one-dimensional convolution kernels (v n-2,1 , h n-2,1 ), (v n-2,2 , h n-2,2 ),..., (v n-2,8 , h n-2,8 ), (v n-1,1 , h n-1,1 ), (v n-1,2 , h n-1,2 ),..., (v n-1,8 , h n-1,8 ), (v n+1,1, h n+1,1 ), (v n+1,2 , h n+1,2 ),..., (v n+1,8 , h n+1,8 ), (v n+2,1 , h n+2,1 ), (v n+2,2 , h n+2,2 ),..., (v n+2,8 , h n+2,8 ), where (v i,j , h i,j ) represents the jth pair of one-dimensional convolution kernels obtained for the i-th frame;
[0040] (2...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


