Video Sub-Bitstream Random Access Using Tile Group Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video bitstream processing methods result in increased coding overhead, network transmission complexity, and access delays due to the need for random access points in sub-bitstreams, particularly in high-resolution videos, leading to inefficient user access and decoding delays.
Innovation Solution
A method and apparatus for processing video bitstreams that determine and configure picture type parameters for random access pictures within sub-bitstreams, allowing for efficient extraction and decoding without relying on decoded information from other pictures, thereby optimizing the distribution of random access points based on user attention and video content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If random access points are added to the entire video bitstream to increase random access capability in high-attention regions, then the random access function is improved, but the coding overhead and processing complexity increase
Solution Approach 1:
The patent divides the video picture into multiple tiles and further segments them into tile groups, allowing independent extraction and random access of specific tile groups without requiring random access points in the entire picture. This segmentation enables region-specific access while reducing overall coding overhead.
Solution Approach 2:
The patent applies different random access strategies to different regions of the video picture based on attention importance. High-attention regions receive denser tile group segmentation for better access capability, while low-attention regions use coarser segmentation, optimizing the balance between access performance and coding efficiency.
2Speed
If random access points are added to the entire video bitstream to enable faster access to attention regions, then the access speed is improved, but the user access delay increases due to waiting for random access pictures
Solution Approach 1:
By segmenting pictures into tiles and tile groups that can be independently decoded and extracted, the patent enables users to access specific region data without waiting for complete random access pictures. This allows faster access to attention regions while avoiding the delay of waiting for entire picture random access points.
Solution Approach 2:
The patent performs preliminary segmentation of the video bitstream into tile and tile group structures during encoding, preparing the data for efficient region-specific extraction. This preliminary organization enables rapid access to specific regions without requiring full picture decoding or waiting for random access pictures.
3Ease of operation
If the identification information field is configured on the entire picture to identify random access capability, then the random access identification is simplified, but the processing complexity of network transmission and server scheduling increases
Solution Approach 1:
The patent extends the identification information field from the entire picture level down to the tile group level within pictures. This hierarchical segmentation allows network devices to identify and process specific tile groups for random access without handling complete picture-level random access structures, reducing processing complexity while maintaining identification capability.
Solution Approach 2:
The patent applies identification information fields selectively to tile groups based on their random access capability and regional importance. This localized application of identification fields optimizes network transmission processing by focusing identification and scheduling resources on specific regions rather than processing entire picture structures.
4Ease of operation
If random access points are distributed throughout the video bitstream to improve access to high-attention regions, then the random access capability is improved, but the terminal cannot access the entire video bitstream when bandwidth is reduced or buffer is full
Solution Approach 1:
The patent segments the video bitstream into independent tile and tile group units that can be selectively decoded and extracted. When bandwidth is limited or buffer is full, terminals can choose to decode and extract only the tile groups corresponding to high-attention regions, achieving partial video content delivery without requiring access to the entire video bitstream.
Solution Approach 2:
The patent enables partial decoding and extraction of tile groups from the video bitstream based on terminal conditions. When resources are constrained, terminals can perform partial action by extracting only necessary tile groups for high-attention regions rather than attempting to access or decode the complete video bitstream, maintaining acceptable video quality under limited conditions.
Data Source
Figure 1~2
Figure 3~4
AI summary
Provided are a method and apparatus for processing a video bitstream, a network device and a readable storage medium. The method includes: parsing a bitstream to be extracted, and acquiring parameter information about an extractable sub-bitstream from a data unit of the bitstream to be extracted; determining a picture type parameter value of a random access picture in the sub-bitstream according to the parameter information; and configuring, in a process of extracting the sub-bitstream from the bitstream to be extracted, a picture type parameter value in a data unit of the sub-bitstream according to the picture type parameter value of the random access picture.