Reserved Bit Prediction for Pipelined Video Slice Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-stage pipelined video encoding, coding units like macroblocks in H.264/AVC or coding tree blocks in H.265/High Efficiency Video Coding are not properly counted in slice size calculation due to encoding delay, leading to inefficient slice size control and potential network congestion.
Innovation Solution
A system and method to predict reserved bits for coding units by calculating bits per coding unit as a function of bitrate, frame rate, and frame size, identifying N reserved bits candidates, and using a fitted function to determine the optimal slice size and header insertion points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If slice size control is implemented in multi-stage pipelined video encoding, then network packet efficiency is improved, but coding units are not properly counted due to encoding delay
Solution Approach 1:
The patent applies preliminary action by estimating the bit consumption of coding units before they are actually encoded. The system calculates predicted bit values for coding units in advance using a prediction model, allowing slice size control to be performed without waiting for the actual encoding to complete. This resolves the timing issue where coding units were not properly counted due to pipeline delays.
2Productivity
If no slice size control is present, then the number of packets per frame is minimized, but packet loss causes significant visual quality degradation
Solution Approach 1:
The patent changes the parameter of slice size by dynamically adjusting it based on predicted bit consumption. By modifying the slice size parameter adaptively, the system achieves a balance between minimizing packet count and maintaining error recovery capability. This allows slice headers to be inserted at optimal positions that prevent decoder disruption while keeping the number of packets low.
3Reliability
If slice headers are inserted frequently, then error recovery capability is improved, but slice efficiency decreases and network congestion increases
Solution Approach 1:
The patent implements feedback by using the prediction model to continuously monitor and adjust slice size based on actual bit consumption patterns. The system learns from encoding results and refines its predictions, allowing it to determine the optimal frequency for inserting slice headers. This feedback mechanism prevents both excessive header insertion (which reduces efficiency) and insufficient header insertion (which harms error recovery).
Data Source
AI summary
A technique to determine an optimum number of reserved bits for use in video encoding. The bits per coding unit is calculated as a function of bitrate, frame rate and frame size. Then for each coding unit, N reserved bits candidates are selected. For each test video the number of packets per frame is identified and the average number of packets per frame is calculated. The best reserved bits candidate out of the N reserved bits candidates is identified, and then a best fitted linear model between best reserved bits candidate and bits per coding unit is determined.


