Virtual Space Emotion Estimation from Gaze Range and User Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in estimating a user's emotion towards a specific object in a virtual space while incurring high calculation loads due to image processing requirements for line-of-sight detection.
Innovation Solution
An information processing device that includes an output control unit, line-of-sight estimation unit, and emotion estimation unit to estimate a user's emotion based on a captured image and a predetermined range on an output image, reducing the need for complex image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If line of sight detection is performed using point light source and image processing on reflected light, then line of sight estimation accuracy is improved, but calculation load increases
Solution Approach 1:
The patent extracts only the essential information needed for line of sight estimation from the captured image, specifically focusing on corneal reflection patterns rather than performing comprehensive image processing. This selective extraction approach maintains estimation accuracy while significantly reducing calculation load by processing only relevant features.
Solution Approach 2:
The patent uses a simplified model or pre-computed data structures to represent line of sight information instead of performing real-time complex image processing. By using pre-established relationships between camera parameters and line of sight vectors, the system achieves accurate estimation without heavy computational burden during operation.
2Measurement precision
If comprehensive image processing is performed to detect line of sight, then line of sight detection accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary calculations and establishes lookup tables or pre-computed data structures that store relationships between camera parameters and line of sight vectors. During actual operation, the system quickly retrieves pre-computed results based on current camera parameters, avoiding time-consuming real-time image processing while maintaining detection accuracy.
Solution Approach 2:
The patent replaces complex mechanical image processing operations with mathematical calculations based on camera geometry and pre-computed data. By substituting intensive image processing algorithms with efficient geometric computations using stored parameter relationships, the system achieves fast line of sight detection without sacrificing accuracy.
Data Source
AI summary
One of the purposes of the present disclosure is to provide a virtual space providing device and the like that are capable of inferring a feeling of a user, who uses a virtual space, toward a specific target while suppressing calculation load. An information processing device according to one aspect of the present disclosure comprises: an output control means for performing control to output an output image, which is an image according to an avatar in a virtual space, to a user who operates the avatar; a line-of-sight inference means for inferring the line of sight of the user on the basis of a predetermined range in the output image; and a feeling inference means for inferring a feeling of the user on the basis of a captured image in which the user imaged by an imaging device is included.