Video Transcoding Using Decoding Data for Faster Predictive Coding

The patent applies preliminary action by performing decoding first to obtain decoding data before encoding. The decoding data obtained from the initial compressed video file is reused as reference information during the encoding process, eliminating the need for redundant processing and reducing hardware resource consumption while improving transcoding speed.

The patent merges the decoding and encoding processes by reusing the decoding data obtained during the initial decoding phase. Instead of treating decoding and encoding as completely separate operations, the method combines them by utilizing the already-decoded information as reference for the encoding process, thereby reducing overall computational load and resource consumption.

The patent implements feedback by using the decoding data obtained during the decoding phase as reference information for the subsequent encoding process. The decoding data is fed back into the encoding process to guide the encoding of coding units, improving transcoding efficiency by leveraging previously processed information without significantly increasing system complexity.

The patent changes the parameter utilization by repurposing the decoding data (originally obtained for decoding purposes) as reference parameters for encoding. Instead of treating decoding and encoding parameters as separate and independent, the method transforms the decoding data into useful reference information for the encoding process, thereby improving efficiency without adding complex new parameters.