Reference Frame Selection for Video Streaming Bandwidth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video conferencing technologies face bandwidth limitations due to high demands for streaming video content, especially in remote working environments with inconsistent connections, and existing compression solutions provide only marginal bandwidth savings at the expense of video quality.
Innovation Solution
An AI/ML-based system that assembles and transmits a set of reference frames covering different facial expressions or object attributes, using a Multi-Layer Perceptron encoder to determine which frames to cache and transmit, and employs an adaptive threshold for blur detection to improve bandwidth efficiency by selectively using blurred frames until a non-blurred frame is available.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If video compression solutions are employed to reduce bandwidth requirements, then bandwidth consumption is reduced, but video quality deteriorates
Solution Approach 1:
The patent segments the video stream into key frames and non-key frames, transmitting only the differences (residuals) between frames rather than complete frames. This segmentation allows bandwidth reduction by sending only change information while maintaining quality through intelligent frame selection and reconstruction at the receiver end.
Solution Approach 2:
The system performs preliminary actions by transmitting reference frames and motion information in advance, allowing the receiver to reconstruct future frames using predicted motion vectors. This preliminary transmission of contextual information enables efficient bandwidth usage while maintaining video quality through predictive coding.
2Loss of energy
If traditional video compression is used, then bandwidth requirements are reduced, but connection quality in remote working environments deteriorates
Solution Approach 1:
The patent implements dynamic adaptation by adjusting the reference frame selection based on real-time network conditions and video content. The system dynamically determines which frames to use as references and how to encode residuals, allowing optimization of both bandwidth efficiency and connection quality under varying network conditions in remote working environments.
Solution Approach 2:
The system incorporates feedback mechanisms where the receiver sends acknowledgment and quality information back to the transmitter, enabling adaptive adjustment of compression parameters and frame selection strategies. This feedback loop ensures connection quality is maintained while optimizing bandwidth usage in remote working scenarios.
3Manufacturing precision
If more frames are transmitted to maintain video quality, then video quality is improved, but bandwidth consumption increases
Solution Approach 1:
The patent extracts only the essential information needed for video reconstruction by separating key frames from non-key frames and transmitting only the residual differences. This extraction approach maintains video quality by preserving critical visual information while reducing bandwidth consumption by omitting redundant data that can be reconstructed from motion vectors and references.
Solution Approach 2:
The system changes encoding parameters dynamically based on frame importance and network conditions. Key frames are transmitted with higher quality parameters while non-key frames use reduced parameters, allowing the system to optimize the balance between video quality and bandwidth consumption by adjusting compression levels on a per-frame basis.
Data Source
AI summary
Systems and methods herein address reference frame selection in video streaming applications using one or more processing units to decode a frame of an encoded video stream that uses an inter-frame depicting an object and an intra-frame depicting the object, the intra-frame being included in a set of intra-frames based at least in part on at least one attribute of the object as depicted in the intra-frame being different from the at least one attribute of the object as depicted in other intra-frames of the set of intra-frames.