Video encoding method and device

Abstract

In the video coding method and device provided by the present application, firstly, key video frames are encoded to obtain encoded images of key video frames; next, the encoded image of the key videoframe is decoded, and the encoded image of the non-key video frame with reduced resolution is obtained by referring to the decoded key video frame image to complete encoding of the video to be encoded. In the above method, based on the human visual line mental model, the complexity of the encoding process is not significantly changed by referring to the key video frames to generate the non-key video frames with reduced resolution, and the reduction of the non-key frame resolution makes the encoding speed increase, and the subjective feeling of video definition is not affected.

Description

technical field [0001] The present application relates to the field of video processing, and in particular, to a video encoding method and device. Background technique [0002] Existing video classification coding methods can be divided into two classes of spatial domain classification and time domain classification. Coding, time-domain coding, MPEG4's FGS (Fine Granular Scalable, fine scalable video coding), H.264's SVC (Scalable video coding, scalable video coding), and wavelet transform coding. [0003] The current video hierarchical coding method has the following defects: [0004] Spatial grading increases the complexity of encoding, and the encoding efficiency of video streams using spatial grading technology is lower than that of video streams without grading; [0005] When time-domain grading is used, the video resolution remains unchanged in the same sequence, and the decoder can discard non-key frames, but the reduction of the frame rate will make the video look ...

AI Technical Summary

Problems solved by technology

[0004] Spatial classification increases the complexity of encoding, and the encoding efficiency of video streams using spatial classification technology is lower than that of video streams without classification;

[0005] When time-domain grading is used, the video resolution remains unchanged in the same sequence, and the decoder can discard non-key frames, but the reduction of the frame rate will make the video look unsmooth and damage the subjective experience;

Images