Layered Video Encoding QP Control for Lower Coding Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding methods face high calculation complexity and low efficiency due to the need to calculate the quantization parameter (QP) of each picture frame based on a configured bit rate control model, which does not account for varying frame complexities and transmission channel conditions.
Innovation Solution
A video coding method that utilizes a layered coding structure to determine quantization parameters based on the frame's layer, using quantization parameter offsets and reference quantization parameters to reduce complexity and improve efficiency, particularly for bidirectionally predicted frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the quantization parameter (QP) of each picture frame is calculated according to a configured bit rate control model, then the bit rate control is achieved, but the calculation complexity increases and video coding efficiency decreases
Solution Approach 1:
The video sequence is divided into multiple picture groups (GPB), each containing multiple picture frames. The QP calculation is segmented so that only the first picture frame in each GPB is calculated using the bit rate control model, while subsequent frames inherit or use simplified QP values, reducing overall calculation complexity while maintaining bit rate control.
Solution Approach 2:
The QP for the first picture frame in each picture group is calculated in advance using the bit rate control model. This preliminary calculation result is then reused for subsequent frames in the same group, avoiding redundant calculations and reducing computational complexity while ensuring bit rate control is established.
2Stability of the object's composition
If the quantization parameter is calculated for each picture frame using bit rate control model, then the bit rate stability is ensured, but the video coding efficiency decreases
Solution Approach 1:
Picture frames are organized into picture groups where only the leading frame undergoes complex QP calculation. This segmentation allows bit rate stability to be established at group level while improving overall coding efficiency by reducing per-frame calculation overhead.
Solution Approach 2:
The QP calculation is performed preliminarily for the first frame of each picture group, establishing the bit rate control parameter in advance. Subsequent frames in the group benefit from this preliminary calculation, improving coding efficiency while maintaining bit rate stability through the inherited QP values.
3Ease of manufacture
If a same quantization offset is obtained for at least two images belonging to different temporal layers, then the encoding simplicity is improved, but the video quality adaptability decreases
Solution Approach 1:
Different quantization offsets are applied to different temporal layers based on their specific quality requirements. Higher temporal layers (requiring higher quality) receive different offset values compared to lower temporal layers, allowing each layer to be optimized locally for its quality requirements while maintaining overall encoding efficiency.
Solution Approach 2:
The quantization offset is made dynamic and adaptable to different temporal layers and picture frame positions within groups. The offset values change based on the temporal layer designation and the frame's position, enabling quality adaptability across different layers while maintaining encoding simplicity through standardized offset application rules.
Data Source
Figure 1A~1B
Figure 2A
Figure 2B~3
AI summary
The present application relates to a video encoding method, a computer device and a storage medium, the method comprising: obtaining a current image frame to be encoded; obtaining the current layer and the image group that the current image frame is located in, wherein the image group includes multiple image frames; obtaining, according to the current layer, a current quantization parameter offset corresponding to the current image frame, wherein image frames of different layers in the image group correspond to different quantization parameter offsets; obtaining a reference quantization parameter corresponding to a reference image frame of the current image frame; obtaining, according to the reference quantization parameter and the current quantization parameter offset, a current quantization parameter corresponding to the current image frame; and encoding the current image frame according to the current quantization parameter.