Pupil Center Mapping in Head-Mounted Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sight positioning systems in VR head-mounted displays require multi-point calibration, which is inefficient and inconvenient, and suffer from reduced accuracy when the user's eye slides relative to the initial calibration position due to the use of absolute pupil center values.
Innovation Solution
A method that determines two transform relations between image, light source, and display unit coordinate systems to accurately map the pupil center from the image coordinate system to the display unit coordinate system, using light spots and transform matrices to convert pupil center coordinates, thereby eliminating the need for calibration and reducing errors caused by eye movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multi-point calibration mode is used for sight positioning, then the model parameters can be solved for current user, but the efficiency and use convenience are low
Solution Approach 1:
The patent extracts the calibration process entirely by using a pre-established transform relationship between coordinate systems. Instead of performing multi-point calibration to solve model parameters, the system directly uses the transform matrix to convert pupil center coordinates from image coordinate system to display coordinate system, eliminating the time-consuming calibration step while maintaining calculation accuracy.
2Measurement precision
If pupil center absolute value is used as model input, then the sight point can be calculated, but when the eye slides relative to the initial calibration state, the sight point error increases and the computed sight point may drift seriously
Solution Approach 1:
The patent introduces a transform matrix as an intermediary between the image coordinate system and the display coordinate system. This transform matrix, established through light spot positioning, serves as a stable reference that accommodates eye movements. By converting pupil center coordinates through this intermediary transform relationship rather than using absolute values directly, the system maintains sight point accuracy even when the eye slides relative to the initial position.
3Measurement precision
If multi-point calibration is performed before each use, then the model parameters are optimized for current user, but the operation complexity and inconvenience increase
Solution Approach 1:
The patent performs the coordinate system transformation setup as a preliminary action during device manufacturing or initial configuration, establishing the transform matrix between image and display coordinate systems. This preliminary setup eliminates the need for repeated multi-point calibration operations before each use, maintaining user-specific accuracy while greatly improving operational convenience.
Data Source
AI summary
A sight positioning method, a head-mounted display device, a computer device, and a computer-readable storage medium are provided. The method includes: determining a first transform relation and a second transform relation, where the first transform relation is a transform relation between an image coordinate system and a light source coordinate system, and the second transform relation is a transform relation between the light source coordinate system and a display unit coordinate system; calculating coordinates of a pupil center in the image coordinate system according to an eye image of eyes of a user acquired by the image acquisition equipment; and obtaining coordinates of the pupil center serving as a fixation point position in the display unit coordinate system of the display unit according to the coordinates of the pupil center in the image coordinate system, the first transform relation and the second transform relation.