Sub Picture Parallel Transcoding for Video Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video devices often struggle to play back video files in unsupported formats or with limited storage capacity, necessitating transcoding to convert files into smaller, supported formats for successful playback and storage.
Innovation Solution
A transcoding system employing multiple encoders that divide video frames into sub-picture partitions and encode them in parallel, using a shared memory to achieve faster-than-real-time transcoding by processing and re-encoding each partition simultaneously.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If video files are transcoded to reduce file size for devices with limited storage capacity, then storage compatibility is improved, but transcoding time increases
Solution Approach 1:
The patent divides each video picture into multiple sub-picture partitions and assigns different partitions to different encoders. This segmentation allows parallel processing of multiple partitions simultaneously, significantly reducing the overall transcoding time while maintaining the file size reduction benefit
Solution Approach 2:
The patent transitions from sequential single-encoder processing to parallel multi-encoder processing by utilizing the dimension of time through simultaneous operation. Multiple encoders work on different sub-picture partitions at the same time, achieving faster-than-real-time transcoding while still reducing file size
2Productivity
If single encoder is used for transcoding to maintain simple system structure, then device complexity is reduced, but transcoding speed decreases
Solution Approach 1:
The system segments the video picture into multiple sub-picture partitions and assigns each partition to a different encoder. This segmentation enables parallel processing architecture that increases transcoding speed while keeping each individual encoder relatively simple
Solution Approach 2:
The patent combines multiple encoders working in parallel to achieve high-speed transcoding. By merging the capabilities of multiple encoders through parallel operation on different sub-picture partitions, the system achieves high productivity without requiring each individual encoder to be overly complex
Data Source
AI summary
A method for transcoding a video file bit stream in parallel at a sub picture latency and faster than real time rate includes writing to a memory a processed raw pixel data stream. The processed raw pixel data stream is expressed as a sequence of pictures. Each picture is divided into a first sub picture partition and a second sub picture partition. A first encoder encodes each first sub picture partition, the first encoder being communicatively coupled to the memory. A second encoder encodes each second sub picture partition, the second encoder being communicatively coupled to the memory.