Gaze Position Detection for Content Recommendation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current information processing apparatuses lack the ability to provide users with personalized content recommendations based on their gaze direction and viewing habits, failing to effectively utilize gaze position data for scene and object identification and metadata analysis.

Innovation Solution

An information processing apparatus that includes a gaze position detection system, which detects and analyzes user gaze on a display screen, generates search information based on focused scenes and objects, and uses this data to recommend content from a storage device, considering priority levels, statistical data from other users, and metadata for improved accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If gaze position detection and analysis functions are added to the information processing apparatus, then content recommendation accuracy and user engagement are improved, but device complexity and processing requirements increase

Engineering Contradiction:
Improvegaze position detection accuracyVSAvoidapparatus structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The information processing apparatus integrates multiple functions including gaze position detection, scene analysis, object recognition, and content recommendation into a single unified system. The control section performs diverse tasks such as detecting gaze positions, analyzing video content, identifying objects, and generating recommendations, making the apparatus multi-functional and reducing the need for separate dedicated devices for each function.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If real-time gaze position detection and content recommendation are implemented, then user engagement and recommendation relevance are improved, but processing time and computational load increase

Engineering Contradiction:
Improvecontent recommendation efficiencyVSAvoidprocessing time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The apparatus performs preliminary analysis of video content by detecting gaze positions and identifying scenes and objects as the user views the content. The control section continuously monitors gaze positions and pre-processes the video data, so when a recommendation is needed, the analysis is already partially complete or can be quickly finalized, reducing the actual recommendation generation time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements a feedback loop where the detected gaze position information is continuously fed back to the control section, which adjusts the content analysis and recommendation generation in real-time. This allows the system to focus computational resources on the most relevant parts of the video content that the user is currently viewing, improving efficiency while maintaining real-time responsiveness.

Inventive Principle:
Principle #23Feedback

3Reliability

If detailed scene and object analysis based on gaze position is performed, then recommendation accuracy is improved, but information processing complexity and data requirements increase

Engineering Contradiction:
Improverecommendation accuracyVSAvoiddata processing burden
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The control section extracts only the essential and relevant information from the video content based on the detected gaze position. Instead of processing all video data, the system identifies and extracts key features such as the current scene, prominent objects in the gaze direction, and relevant metadata, discarding unnecessary information and reducing the data processing burden while maintaining recommendation accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The apparatus applies different levels of analysis depth to different parts of the video content based on local relevance. Areas or objects that are currently in the user's gaze receive detailed analysis, while other parts receive minimal or no processing. This localized quality approach ensures high accuracy for relevant content while reducing overall processing complexity and data requirements.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3040878B1Information processing device and information processing method
Publication Date: 2019.10.02 SONY GROUP CORP
  • EP3040878B1 patent drawingFigure 1
  • EP3040878B1 patent drawingFigure 2
  • EP3040878B1 patent drawingFigure 3

AI summary

[Object] [Solving Means] An information processing apparatus includes: a gaze position detection section that detects a gaze position of a user with respect to a display screen; a position information acquisition section that acquires position information of the user; and a control section that judges, based on the detected gaze position, a scene and an object in the scene that the user focuses in a content displayed on the display screen, and generates search information from the judged object and the acquired position information.