Media Metadata Encoding Overhead Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing MP4 file streaming technology faces significant bandwidth overhead due to metadata transmission, particularly in lower bit rate scenarios, where much of the metadata range is unused, leading to inefficiencies.
Innovation Solution
Implementing metadata optimizations such as reducing the size of metadata fields, consolidating information, and omitting optional parameters, along with selecting appropriate timescales and network abstraction layer units, to minimize metadata overhead. These optimizations include using 2-byte fields for image size representation, uniform sample durations, consolidating subsample mappings, and omitting unnecessary NALUs, especially for lower bit rate scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard MP4 metadata fields are used for streaming, then compatibility and completeness of information are maintained, but bandwidth overhead increases significantly (10-15 kbps)
Solution Approach 1:
The patent extracts and removes redundant metadata fields from the MP4 stream, keeping only the essential information needed for playback. This includes removing optional parameters and consolidating information that can be inferred from other parts of the stream, thereby reducing bandwidth overhead while maintaining reliability.
Solution Approach 2:
The patent merges multiple metadata fields into consolidated representations where possible. For example, it combines timing information and sample duration data into more efficient representations, and merges redundant parameter sets that appear multiple times in the metadata hierarchy, reducing overall metadata size.
2Adaptability or versatility
If 4-byte fields are used for image size representation in metadata, then maximum image sizes up to 4 gigabytes can be represented, but bandwidth is wasted in lower bit rate scenarios where images are smaller
Solution Approach 1:
The patent applies local quality by using different field sizes for different parts of the metadata based on actual content requirements. For lower bit rate scenarios where image sizes are small, it uses compact 2-byte fields. For higher bit rate scenarios or when large images are present, it uses full 4-byte fields, optimizing bandwidth usage while maintaining adaptability.
Solution Approach 2:
The patent changes the parameter representation from fixed 4-byte fields to variable-length fields that adapt to the actual data requirements. It implements this by using 2-byte fields for image size when sufficient, and only resorting to 4-byte fields when larger values are needed, thereby reducing bandwidth waste while maintaining versatility.
3Loss of energy
If uniform sample durations are implemented, then metadata overhead is reduced and streaming efficiency improves, but flexibility in handling variable frame rates is limited
Solution Approach 1:
The patent implements dynamics by allowing the sample duration representation to adapt between uniform and variable based on the content. It uses uniform sample durations as the default to minimize overhead, but introduces mechanisms to handle variable frame rates when necessary, balancing metadata efficiency with frame rate flexibility.
Solution Approach 2:
The patent segments the stream into sections with uniform sample durations and sections with variable durations, using segmentation boxes to delimit these regions. This allows the majority of the stream to benefit from uniform duration efficiency while accommodating variable frame rate segments where needed.
4Reliability
If all NALUs are included in the metadata, then complete video information is preserved, but unnecessary NALUs increase metadata size and bandwidth consumption
Solution Approach 1:
The patent extracts and removes unnecessary NALUs from the metadata, keeping only those that are essential for video decoding and playback. It identifies and eliminates redundant parameter sets and NAL units that do not contribute to actual video quality or decoding requirements, reducing metadata volume while preserving completeness.
Data Source
AI summary
Techniques for encoding media content are described that reduce the overhead associated with metadata when the encoded content is streamed over a network. Some of these techniques involve the selection of parameters or consolidation of information to ensure consistency or uniformity of metadata values. Others relate to the recognition and removal of redundancies or optional information. Still others take advantage of characteristics of specific bit rate scenarios to more efficiently represent information.