Scalable Video Encoder Layer-Specific Buffer Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scalable video coding technologies inefficiently manage decoded picture buffers, leading to excessive memory usage and initial playback delays when decoding subsets of temporal scalable bitstreams, as they lack optimal buffer size allocation and picture reordering information.
Innovation Solution
An encoder generates and embeds information about the minimum and maximum picture buffer sizes and reordering sizes for each layer within the encoded scalable data stream, using signaling protocols like Session Initiation Protocol or Real-Time Streaming Protocol, to enable efficient memory allocation and reduced playback delays in decoders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If current scalable video coding technologies are used to decode subsets of temporal scalable bitstreams, then decoding functionality is achieved, but excessive memory usage occurs in decoded picture buffers
Solution Approach 1:
The patent segments the decoded picture buffer management into layer-specific components. Each layer (base layer and enhancement layers) has its own minimum and maximum picture buffer sizes and reordering sizes defined in the bitstream. The decoder allocates memory separately for each layer based on these parameters, allowing efficient memory management when decoding subsets of layers. This segmentation enables the decoder to allocate only the necessary memory for the decoded layers rather than provisioning for the entire scalable bitstream.
2Loss of time
If current scalable video coding technologies are used to decode subsets of temporal scalable bitstreams, then decoding functionality is achieved, but initial playback delays occur
Solution Approach 1:
The patent applies preliminary action by defining minimum and maximum picture buffer sizes and reordering sizes in advance for each layer during encoding. These parameters are signaled in the bitstream so the decoder can pre-allocate the exact buffer sizes needed before playback begins. This eliminates the need for the decoder to wait or dynamically adjust buffer sizes during playback, thereby reducing initial playback delays while ensuring smooth playback operation.
3Adaptability or versatility
If buffer sizes are allocated for the entire temporal scalable bitstream, then all layers can be decoded, but memory is wasted when only subsets are decoded
Solution Approach 1:
The patent implements dynamic memory allocation by allowing the decoder to adapt buffer sizes based on the actual layers being decoded. The minimum and maximum picture buffer sizes and reordering sizes are defined per layer in the bitstream, enabling the decoder to dynamically adjust memory allocation according to the subset of layers being decoded. This dynamic approach maintains adaptability to decode any layer subset while optimizing memory usage to match actual decoding requirements.
4Speed
If conventional B picture concepts are used for temporal scalability, then frame rate reduction is achieved, but reference picture management becomes complex
Solution Approach 1:
The patent changes the approach from conventional B picture concepts to a parameter-based buffer management system. Instead of using B pictures with complex bi-directional prediction and reference picture marking, the patent defines minimum and maximum picture buffer sizes and reordering sizes as explicit parameters in the bitstream for each layer. This parameter change simplifies reference picture management while maintaining temporal scalability and frame rate control capabilities.
Data Source
AI summary
An encoder for encoding a video signal, wherein the encoder is configured to generate an encoded scalable data stream comprising a base layer and at least one enhancement layer, wherein the encoder is further configured to generate information associated with each of the base layer and the at least one enhancement layer.