A video image interpolation optimization method and filter

Abstract

The present application discloses a video image interpolation optimization method, which includes: obtaining sequence information of video data sequence; determining the interpolation filter used according to the sequence information; determining the pixel precision, and determining the inserted component according to the pixel precision The number of pixels; using the interpolation filter to calculate the sub-pixels according to a predetermined rule. The application also discloses a filter. In the specific implementation of the present application, since the interpolation filter used is first determined according to the sequence information; then the pixel precision is determined, and the number of sub-pixel points to be inserted is determined according to the pixel precision, and the selected interpolation filter is used according to the predetermined rule Calculate the sub-pixels. The present application can select the number of taps of the interpolation filter according to the sequence information, and insert sub-pixel points, which improves the performance of encoding and decoding, and reduces the complexity of interpolation calculation.

Description

technical field [0001] The present application relates to the field of video image processing, in particular to a video image interpolation optimization method and filter. Background technique [0002] With the gradual popularization of many high-definition digital products such as digital TV, Internet high-definition video, digital camera, digital video camera, etc., the existing video codec standards can no longer meet the requirements, especially for high-resolution video, such as 4K×2K very high resolution. Therefore, people look forward to the development of a new generation of high-performance video coding standards that can adapt to ultra-high-definition video. [0003] In the video coding and decoding technology, the sub-pixel interpolation technology can improve the accuracy of motion prediction, which has a very positive effect on the improvement of the entire coding and decoding performance. Sub-pixel interpolation was first used in MPEG-1, but it is limited to ...

AI Technical Summary

Problems solved by technology

Sub-pixel interpolation was first used in MPEG-1, but it is limited to half-pixel motion estimation. Although it is better than full-pixel motion estimation, the accuracy and performance are not high; starting from MPEG-4ASP, the interpolation accuracy has been improved, using 1 / 4 Pixel-accurate interpolation; H.264 / AVC uses a 6-tap filter to obtain 1 / 2 pixels, and then uses an average filter to obtain 1 / 4 pixels, the performance has been greatly improved; it is also used in AVS Jizhun Profile A 4-tap filter is used to obtain 1 / 2 pixels, and then 1 / 4 pixels; in the recent JCT-VC (the Joint Collaborative Team on Video Coding) is developing the next generation of high-performance video coding standard HEVC (High Efficiency In Video Coding), in order to further improve the interpolation performance, a high-performance separable DCT-based interpolation filter is used, which reduces the bit rate a lot compared to H.264 / AVC and can significantly improve the coding performance, but its interpolation calculation higher complexity

Images