Picture Encoding Device Using Offset Prediction Quantization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing picture encoding techniques face inefficiencies in reducing the code amount of quantization parameters, particularly when using adaptive quantization, leading to increased code amounts due to suboptimal prediction methods.
Innovation Solution
A picture encoding device and method that derive an optimum prediction quantization parameter from preceding blocks, calculate the difference with the actual quantization parameter, and encode this difference, allowing for efficient code amount control by predicting quantization parameters using multiple previous blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the quantization parameter of the previous block is used as the prediction quantization parameter, then the code amount is reduced, but when adaptive quantization is performed, the difference quantization parameter becomes large and the code amount increases
Solution Approach 1:
The patent segments the prediction process by dividing the prediction quantization parameter into two components: a base value from the previous block and an offset value derived from activity information. This segmentation allows the system to maintain the simplicity of using previous block parameters while adding a correction term to improve prediction accuracy for adaptive quantization scenarios.
Solution Approach 2:
The patent changes the prediction parameter by introducing an offset value that is added to the base quantization parameter. This offset is calculated based on activity information (such as variance or gradient) of the current block, allowing the prediction to adapt to local picture characteristics and reducing the difference quantization parameter.
2Manufacturing precision
If adaptive quantization is performed to improve subjective picture quality, then picture quality is improved, but the difference quantization parameter becomes large and code amount increases
Solution Approach 1:
The patent implements feedback by using activity information (variance, gradient, or edge detection results) of the current block to calculate an offset value. This feedback mechanism allows the prediction quantization parameter to adapt to local picture characteristics, improving prediction accuracy and reducing the code amount of difference quantization parameters while maintaining picture quality.
Solution Approach 2:
The patent dynamically changes the prediction parameter by adding an offset value derived from activity information to the base quantization parameter. This parameter change enables the system to achieve better prediction accuracy for adaptive quantization without significantly increasing the code amount, thus resolving the contradiction between picture quality and code efficiency.
3Ease of manufacture
If the quantization parameter of the previous block is used for prediction, then encoding simplicity is maintained, but prediction accuracy deteriorates leading to larger difference quantization parameters
Solution Approach 1:
The patent segments the prediction process into a simple base component (previous block parameter) and a correction component (offset based on activity information). This segmentation maintains encoding simplicity by keeping the base prediction straightforward while adding a relatively simple offset calculation to improve accuracy.
Solution Approach 2:
The patent modifies the prediction parameter by adding a calculated offset to the base quantization parameter. This parameter change improves prediction accuracy without significantly complicating the encoding process, as the offset can be derived from readily available activity information such as variance or gradient calculations.
Data Source
AI summary
There is provided a picture encoding device that encodes a picture and encodes a difference quantization parameter in a unit of a quantization coding block which is divided from the picture and is a management unit of a quantization parameter. A quantization parameter calculator derives a quantization parameter of the quantization coding block to be encoded. A prediction quantization parameter derivation unit derives a prediction quantization parameter using the quantization parameters of a plurality of quantization coding blocks which precede the quantization coding block to be encoded in order of encoding. A difference quantization parameter generator derives a difference quantization parameter of the quantization coding block to be encoded, using a difference between the quantization parameter of the quantization coding block to be encoded and the prediction quantization parameter. A first bitstream generator encodes the difference quantization parameter of the quantization coding block to be encoded.