Scalable Video Coding Extension for Adaptive Resolution Change
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding systems lack flexibility in adapting to varying resolution requirements, especially in environments where endpoints have different capabilities, and struggle with efficient encoding and decoding of scalable video content, particularly in 3D video applications.
Innovation Solution
The implementation of a scalable video coding extension that allows adaptive resolution change by indicating specific picture types in the bitstream to utilize inter-layer prediction, enabling each access unit to contain a single picture from a single layer and using inter-layer scalability tools for flexible bitstream trimming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If scalable video coding extension is implemented to enable adaptive resolution change, then adaptability to different endpoint capabilities is improved, but device complexity increases
Solution Approach 1:
The video bitstream is segmented into multiple layers (base layer and enhancement layers) with different resolutions. The decoder can selectively decode only the base layer or both base and enhancement layers depending on endpoint capabilities, enabling adaptive resolution change without requiring the entire scalable structure to be processed at full complexity
Solution Approach 2:
The system dynamically adapts the decoding process based on endpoint capabilities. The decoder can switch between decoding only the base layer (lower resolution) and decoding both base and enhancement layers (higher resolution) by utilizing the layer_id and nuh_layer_id parameters to identify and process appropriate layers
2Productivity
If inter-layer prediction is used for all picture types in enhancement layer, then coding efficiency is improved, but decoding complexity increases
Solution Approach 1:
Inter-layer prediction is applied selectively based on picture type and layer characteristics. The patent indicates that only certain types of pictures (e.g., RAP pictures or specific NAL unit types) in the enhancement layer utilize inter-layer prediction, while other picture types may use intra-layer prediction or no prediction. This localized application of prediction techniques optimizes coding efficiency while controlling decoding complexity
3Adaptability or versatility
If each access unit contains multiple pictures from multiple layers, then scalability is improved, but processing complexity increases
Solution Approach 1:
The patent extracts and separates pictures from different layers into distinct access units. Each access unit is configured to contain a single picture from a single layer (which may or may not be a base-layer picture), except for switching pictures. This extraction simplifies the processing structure while maintaining scalability through the use of layer_id and nuh_layer_id parameters for layer identification
Data Source
Figure 1~2
Figure 3
Figure 4a~4b
AI summary
Various methods, apparatuses and computer program products for video encoding and decoding. In some embodiments a first picture having a first resolution on a first layer is encoded without inter-layer prediction; a second picture having a second resolution on a second layer is encoded without inter-layer prediction; a third picture having the second resolution on the second layer is encoded without inter-layer prediction and without temporally referencing to pictures prior to the second picture. Further, a corresponding indication is encoded. In some embodiments one or more indications are received to determine if a switching point, referring to a picture coded using inter- layer prediction, from the first layer to the second layer exists, and if so, pictures in the first layer prior to the switching point; pictures associated with the switching point; and pictures in the second layer after the switching point may be decoded.