Variable-Quality Video Streaming for High-Relevance Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video streaming applications do not support the intelligent identification and buffering of high-relevance regions within video frames, leading to degraded viewing experiences and inefficient data usage when users are forced to choose between high-quality videos consuming large amounts of internet data or low-quality videos lacking detail.
Innovation Solution
Implementing a convolutional neural network (CNN) model to identify high-relevance regions within video frames and buffer them at a minimum frame quality while maintaining lower quality for other regions, allowing for internally-variable frame quality streaming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If uniform high-quality video frames are streamed, then viewing experience is improved, but data consumption increases excessively
Solution Approach 1:
The patent applies local quality by differentiating video frame regions into high-relevance and low-relevance areas using CNN analysis. High-relevance regions (containing important visual information) are streamed at higher quality, while low-relevance regions are streamed at lower quality, thereby maintaining viewing experience for critical content while reducing overall data consumption.
Solution Approach 2:
The patent segments the video frame into multiple quality zones based on region importance. By dividing the frame into high-relevance and low-relevance regions and applying different quality levels to each segment, the system optimizes the balance between viewing experience and data usage efficiency.
2Loss of energy
If uniform low-quality video frames are streamed, then data consumption is reduced, but viewing experience degrades
Solution Approach 1:
The patent ensures that even when overall quality is reduced, high-relevance regions maintain adequate quality levels necessary for acceptable viewing experience. The CNN-based region analysis identifies critical areas that must retain sufficient quality, while only non-critical regions receive reduced quality treatment.
3Loss of energy
If frame quality is dynamically adjusted by region, then data efficiency is improved, but system complexity increases
Solution Approach 1:
The patent introduces a CNN-based region analysis intermediary that automatically identifies high-relevance regions. This intermediary component handles the complex task of quality determination, allowing the rest of the streaming system to operate with simpler logic while achieving sophisticated region-based quality optimization.
Solution Approach 2:
The system employs self-service through automated CNN-based region classification and quality determination. The algorithm autonomously analyzes video content, identifies important regions, and assigns appropriate quality levels without requiring manual intervention or complex external control systems.
Data Source
AI summary
A method for streaming videos with internally-variable frame quality is implemented via a computing system including a processor. The method includes accessing a video including and evenly sampling the video frames. The method includes, for each sampled video frame, analyzing the video frame using a CNN model to determine whether the video frame includes high-relevance region(s). The method also includes, for each sampled video frame including high-relevance region(s), extracting coordinates of the high-relevance region(s) using the CNN model, and for each sampled video frame including high-relevance region(s) and each intervening video frame between the sampled video frame and a next sampled video frame, setting a minimum frame quality for the extracted coordinates. The method further includes streaming the video with internally-variable frame quality by buffering the extracted coordinates within the video frames with the minimum frame quality, while buffering remaining coordinates within the video frames with a lower frame quality.