Image Coding Interlayer Prediction Reference Layer Designation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing image coding systems, such as those proposed in Non-Patent Documents 2 and 3, experience a decrease in coding efficiency due to the unnecessary transmission of information related to interlayer prediction in the slice header, leading to increased coding amounts and reduced efficiency.
Innovation Solution
An image coding apparatus and method that generates and transmits reference layer designation information only when necessary, based on the specific conditions of interlayer prediction, to avoid unnecessary syntax transmission in the slice header, thereby controlling the interlayer prediction process effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reference layer designation information is always transmitted in the slice header for interlayer prediction, then the interlayer prediction can be performed, but the coding amount of the slice header increases unnecessarily
Solution Approach 1:
The patent extracts only the necessary reference layer designation information from the slice header by checking whether the maximum number of reference layers coincides with the actual number of reference layers. When they coincide, the reference layer designation information is omitted, thereby removing unnecessary data transmission while preserving interlayer prediction functionality.
Solution Approach 2:
The patent implements dynamic transmission of reference layer designation information based on the actual prediction requirements. The system adapts the slice header content by conditionally including or excluding reference layer designation information according to whether the maximum number of reference layers matches the actual number, making the header structure flexible rather than static.
2Loss of information
If unnecessary syntax is transmitted in the slice header, then the interlayer prediction information is complete, but the coding efficiency decreases
Solution Approach 1:
The patent removes unnecessary syntax elements from the slice header by comparing the maximum number of reference layers with the actual number of reference layers. When these values coincide, the reference layer designation information is extracted and omitted from transmission, thereby eliminating redundant data while maintaining complete interlayer prediction information.
Solution Approach 2:
The patent changes the transmission parameter of reference layer designation information from a fixed state to a conditional state. By monitoring whether the maximum number of reference layers coincides with the actual number, the system dynamically adjusts the inclusion of this parameter in the slice header, optimizing the balance between information completeness and coding efficiency.
3Adaptability or versatility
If the maximum number of reference layers and the actual number of reference layers are both set for the whole image data, then the interlayer prediction can be controlled, but the coding amount increases when they do not coincide
Solution Approach 1:
The patent extracts and omits the reference layer designation information from the slice header when the maximum number of reference layers coincides with the actual number. This selective extraction approach maintains the adaptability to control interlayer prediction through the preset maximum number while removing the redundant designation information that would otherwise increase the coding amount.
Solution Approach 2:
The patent makes the maximum number of reference layers parameter serve multiple functions: it both controls the interlayer prediction capability and acts as a default reference layer designation when it coincides with the actual number. This multi-functionality eliminates the need for separate reference layer designation information in many cases, reducing the coding amount while preserving control flexibility.
Data Source
AI summary
An image coding apparatus and method capable of suppressing a decrease in coding efficiency. The image coding apparatus includes: a coding unit that codes image data configured by plural layers; an interlayer prediction-related information generating unit that generates information relating to an interlayer prediction of the current slice in a case a layer of the current slice is an enhancement layer, and a slice type of current slice is a P slice or a B slice; and a transmission unit that transmits coded data of the image data generated by the coding unit and the information relating to the interlayer prediction generated by the interlayer prediction-related information generating unit. The apparatus and method, for example, may be applied to an image coding apparatus that scalably codes image data or an image decoding apparatus that decodes coded data acquired by performing scalable coding of the image data.