Streaming Service Image Segmentation ROI Bandwidth

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing streaming technologies, including MPEG-DASH, are insufficient for seamless streaming of high-resolution video content due to bandwidth limitations and quality of service issues, leading to buffering and interruptions.

Innovation Solution

A method and apparatus for providing a streaming service based on image segmentation, where an electronic device requests and parses a media presentation description file, detects region-of-interest (ROI) in images, and transmits ROI-specific high-resolution segments along with lower-resolution background images, allowing for seamless playback without interruptions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If MPEG-DASH streaming technology is used, then adaptability to network environment is improved, but seamless streaming of high-resolution video content cannot be achieved due to bandwidth limitations

Engineering Contradiction:
Improveadaptability to network environmentVSAvoidseamless streaming capability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The video content is segmented into multiple image segments at different resolutions (e.g., first resolution for background, second resolution for ROI). This segmentation allows selective transmission of only necessary high-resolution segments, reducing overall bandwidth requirements while maintaining seamless streaming capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the video content are assigned different quality levels based on their importance. Regions of interest (ROI) are transmitted at higher resolution while background areas are transmitted at lower resolution. This local quality differentiation maintains visual quality where needed while reducing overall bandwidth consumption, enabling seamless high-resolution streaming.

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If high-resolution video content is transmitted, then video quality is improved, but bandwidth requirements increase causing buffering and interruptions

Engineering Contradiction:
Improvevideo qualityVSAvoidbandwidth requirements
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

High-resolution video content is divided into multiple image segments, each representing a specific region. Only segments corresponding to regions of interest are transmitted at full high resolution, while other segments are transmitted at lower resolution or skipped. This reduces the total quantity of data transmitted while preserving video quality in important areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The transmission system applies different quality levels to different spatial regions of the video content. High-resolution transmission is applied locally to regions of interest (such as moving objects or important scenes), while other regions receive lower resolution transmission. This maintains video quality where needed while significantly reducing overall bandwidth requirements.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS10924782B2Method of providing streaming service based on image segmentation and electronic device supporting the same
Publication Date: 2021.02.16 KOREA ELECTRONICS TECH INST
  • US10924782B2 patent drawing
  • US10924782B2 patent drawing
  • US10924782B2 patent drawing

AI summary

The present disclosure provides a system for providing an image-segmentation based streaming service. The system includes a communication circuit configured to establish a communication channel with a streaming server, a storage configured to store data provided by the streaming server, a display configured to output an image, a processor functionally connected to the communication circuit, the storage, and the display. The processor can detect a region-of-interest (ROI) of an output image associated with a media presentation description (MPD) provided by the streaming server and transmit the ROI information to the streaming server. The processor can also receive image segments of a first resolution corresponding to the ROI information and a full image of a second resolution from the streaming server. The processor can also synthesize the image segments of the first resolution and the full image of the second resolution to output a synthesized reconstruction image through the display.