360-Degree Video ROI Signaling for Efficient Viewport Rendering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding techniques struggle to efficiently compress and transmit 360-degree video data, particularly in high-fidelity formats, due to the large amount of data required, which burdens communication networks and devices, and lack effective methods for signaling regions of interest (ROIs) within the video content.

Innovation Solution

The system generates and processes media files for 360-degree video content by including signaling information for regions of interest (ROIs) in three-dimensional spherical and two-dimensional projected spaces, facilitating both transmission and rendering of the video data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If video data is compressed using conventional video coding techniques, then data transmission efficiency is improved, but video quality and fidelity deteriorate

Engineering Contradiction:
Improvedata transmission efficiencyVSAvoidvideo quality and fidelity
Core Design Contradiction:
Loss of energyVSManufacturing precision

Solution Approach 1:

The patent segments the 360-degree video content into multiple regions of interest (ROIs) and processes them differently. Full-resolution data is transmitted for identified ROIs while other regions use compressed representations, resolving the contradiction between compression efficiency and quality preservation by applying different compression levels to different spatial segments

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements local quality enhancement by maintaining high fidelity specifically in regions of interest while allowing lower quality in non-critical areas. This is achieved through selective transmission of full-resolution data for ROIs identified through metadata analysis, viewer behavior prediction, or importance mapping, thus preserving video quality where it matters most while improving overall transmission efficiency

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If high-fidelity video data is transmitted to meet consumer demands, then video quality is improved, but network burden and device processing load increase

Engineering Contradiction:
Improvevideo qualityVSAvoidnetwork burden and device processing load
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent extracts and transmits only the essential high-fidelity data for regions of interest rather than transmitting complete high-resolution 360-degree video. By identifying and isolating ROIs through metadata, viewer behavior analysis, or importance mapping, the system extracts minimal necessary data at full quality while compressing or omitting less critical regions, thereby reducing network burden and device processing requirements

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by transmitting full-resolution data only for specific regions of interest rather than for the entire 360-degree video content. This selective approach provides high quality where needed (excessive action in critical areas) while using compression for other regions, balancing video quality requirements with reduced network and device loads

Inventive Principle:
Principle #16Partial or excessive action

3Manufacturing precision

If complete 360-degree video data is processed, then rendering quality is improved, but processing time and energy consumption increase

Engineering Contradiction:
Improverendering qualityVSAvoidprocessing time and energy consumption
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary identification and marking of regions of interest during video encoding or pre-processing, storing ROI metadata that guides subsequent rendering operations. This preliminary action allows the rendering device to quickly locate and process only critical high-fidelity regions without analyzing the entire 360-degree video data, significantly reducing processing time and energy consumption while maintaining rendering quality for important areas

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent extracts and prioritizes processing of region of interest data while deferring or simplifying processing of non-critical regions. By separating ROI identification from full video processing and using metadata to guide rendering, the system extracts essential high-quality data that needs immediate processing while allowing less critical regions to be handled with reduced resources, thereby reducing overall processing time and energy consumption

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3552394B1Systems and methods of signaling of regions of interest
Publication Date: 2025.11.19 QUALCOMM INC
  • EP3552394B1 patent drawingFigure 1
  • EP3552394B1 patent drawingFigure 2A~2B
  • EP3552394B1 patent drawingFigure 3

AI summary

Techniques and systems are provided for processing video data. In one example, a media file associated with 360-degree video data can be obtained. The 360-degree video data may include a spherical representation of a scene. The media file may include first signaling information and second signaling information of a viewport region corresponding to a region of interest (ROI) in the spherical representation. The first signaling information may include a center position and a dimension of the viewport region measured in a spherical space associated with the spherical representation. The second signaling information may indicate a region of a picture comprising the viewport region, the picture being formed by projecting the spherical representation including the ROI onto a plane. Pixels corresponding to the viewport region from the data of the picture can be extracted based on the first signaling information and second signaling information, and can be provided for rendering.