An Efficient and Adaptive Video Transcoding System Based on Data Mining

Abstract

The invention discloses an efficient self-adaptive video transcoding system and method based on data mining. The method includes the following steps that step1, a video, to be transcoded, of a first compression standard is input; step2, the video to be transcoded is decoded to generate a video to be encoded, and decoding information of all macro blocks of all video frames is extracted; step3, the extracted decoding information is input in an encoding macro block mode decision-making tree, and encoding macro block modes for all the micro blocks of all the video frames of the video to be encoded are determined through the encoding macro block mode decision-making tree; step4, according to the determined encoding macro block modes and motion vectors of all the micro blocks, all the micro blocks of all the video frames of the video to be encoded are encoded according to a second compression standard, so that the video to be encoded is encoded to be a transcoded video of the second compression standard.

Description

technical field [0001] The invention belongs to the field of video coding and video transcoding, and in particular relates to an efficient and adaptive video transcoding system based on data mining, which can decode and re-encode videos of different encoding formats, different code rates, and different resolutions, thereby Transcode to a video that meets the required encoding format, bit rate, and resolution. Background technique [0002] Technical solution of prior art one [0003] What is implemented in prior art 1 is the low-complexity video transcoding technology of MPEG-2 / H.263 → H.264 at the same resolution, which uses machine learning to find the decoding information of MPEG-2 / H.263 The connection with the H.264 encoding information establishes a fast encoding mechanism at the encoding end and realizes the rapid transcoding process. Prior Art One assumes that there is a certain correlation between the CBP, encoding mode, mean and variance of the MPEG-2 / H.263 macrobl...

AI Technical Summary

Problems solved by technology

[0007] 2. Only use the decision tree model with a single quantization parameter (QP=25) to realize the transcoding process. In actual transcoding, the QP of each macroblock in each video frame is different. Using the decision tree model with QP=25 is for The transcoding effect of the video to be transcoded with QP near 25 is better, but the transcoding effect of the QP video farther away from 25 is not good

[0013] 1. Only node 1 and node 3 use the Jrip classifier, node 2 and node 4 use the original H.264 full-decoding and full-editing technology, so further improvement is needed

[0014] 2. Only use a decision tree with a single quantization parameter (QP=25) to implement the transcoding process. The transcoding effect of the video to be transcoded with a QP near 25 is better, but the QP of the video to be transcoded deviates far from 25. There will be underfitting (bad effect), so a single QP decision tree is not suitable for a wide range of applications

Images