Adaptive Video Coding Offloading for Density-Quality Tradeoffs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding systems face challenges in achieving optimal efficiency in terms of compression, quality, and computational complexity across various coding contexts, particularly with the increasing complexity of newer codecs like AV1 and HEVC.
Innovation Solution
The system employs an adaptive hybrid approach by selectively offloading video coding processes between hardware and software based on target video quality levels, latency requirements, and the functional capabilities of the hardware and software, using an optimization algorithm to achieve the best speed/latency/quality tradeoffs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If hardware encoding is used, then processing speed and latency are improved, but video quality and compression efficiency deteriorate
Solution Approach 1:
The video encoding process is divided into multiple independent encoding passes (e.g., first pass, second pass, third pass) with different complexity levels. Hardware acceleration is applied selectively to specific passes or portions of the encoding process, allowing high-speed processing for non-critical segments while reserving software-based high-quality encoding for critical segments that require superior video quality.
Solution Approach 2:
Different quality levels are applied to different portions or passes of the video encoding process. Instead of uniformly applying high-quality software encoding to the entire video stream, the system applies high-quality encoding only to specific regions, passes, or frames where quality is most critical, while using faster hardware encoding for other portions.
2Manufacturing precision
If software encoding is used, then video quality and compression efficiency are improved, but computational complexity and processing time increase
Solution Approach 1:
The system dynamically adjusts the encoding strategy by selecting between hardware and software encoding based on real-time conditions such as target video quality levels, latency requirements, and available computational resources. This dynamic selection allows the system to optimize the balance between quality and complexity for each encoding task.
Solution Approach 2:
The system changes encoding parameters such as complexity level, pass number, and quality target to match the available computational resources and desired output quality. By adjusting these parameters, the system can reduce computational complexity when hardware acceleration is used while maintaining acceptable video quality through parameter optimization.
3Manufacturing precision
If multiple encoder passes and mode searches are executed, then video quality is improved, but processing time and latency increase
Solution Approach 1:
Instead of always executing the full set of encoder passes and mode searches, the system applies partial encoding actions by selecting only the necessary number of passes and search depths required to achieve the target quality level. This partial action approach reduces processing time while maintaining sufficient video quality for the given application requirements.
Data Source
AI summary
Techniques related to distributing the video encoding processing of an input video across hardware and software systems. Such techniques include evaluating the content of the video and determine whether or the encoding operation is best to be done on the hardware system only, software system only or a hybrid hardware and software system.