HMD Image Rotation for Head-Camera Roll Misalignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional head-mounted display rendering methods fail to adequately address motion cyber sickness, particularly when the user's head roll motion does not align with the roll angle of the image capturing device during image acquisition, leading to discomfort and potential sickness.
Innovation Solution
A method that accounts for both the user's current head roll angle and the roll angle of the image capturing device at the time of image acquisition, adjusting the image rendering accordingly by activating a detection flag and applying a rotational angle to align the user's head with the image capturing device's inclination, while providing a graphical indicator for user feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional rendering methods compensate only for user's head roll motion, then the rendering is simple and fast, but motion cyber sickness occurs when there is misalignment between visual perception and vestibular system perception
Solution Approach 1:
The patent applies preliminary action by pre-obtaining the capturing roll angle metadata from the image data before rendering. This metadata is stored and ready for use during the rendering process, allowing the system to compensate for both head roll motion and camera roll angle without adding computational complexity during real-time rendering. The detection flag is also set in advance based on threshold comparison, enabling proactive adjustment of the rotational angle application.
Solution Approach 2:
The patent introduces an intermediary mechanism through the detection flag that mediates between the simple conventional rendering approach and the more complex dual-roll compensation approach. The detection flag acts as a switch that activates additional rotational angle compensation only when needed (when capturing roll angle indicates significant camera tilt), allowing the system to maintain simplicity in normal cases while preventing motion sickness when necessary.
2Stability of the object's composition
If the renderer applies reverse rotation to keep the horizon horizontal during head tilting, then visual consistency is maintained, but discomfort increases when the capturing device had a non-zero roll angle during acquisition
Solution Approach 1:
The patent applies local quality by making the rotational angle compensation adaptive rather than uniform. When the detection flag is not activated (capturing roll angle is small), only the standard head roll compensation is applied. When the detection flag is activated (capturing roll angle is significant), an additional rotational angle equal to the capturing roll angle is applied. This localized adjustment ensures visual consistency is maintained in normal cases while preventing discomfort in cases with significant camera tilt.
Solution Approach 2:
The patent changes the rotational angle parameter dynamically based on the detection flag state. The rotational angle is set to one value when the detection flag is not activated and to a different value (capturing roll angle) when the detection flag is activated. This parameter change allows the system to adapt the rendering behavior to match the capturing conditions, resolving the conflict between visual consistency and user comfort.
3Object-affected harmful factors
If the system increases processing to account for both head roll and camera roll angles, then motion sickness is reduced, but processing time and computational load increase
Solution Approach 1:
The patent minimizes processing time by performing the capturing roll angle comparison and detection flag setting in advance, before the actual rendering occurs. The metadata is extracted and the detection flag is set during image acquisition or pre-processing, so that during real-time rendering, the system only needs to apply a pre-determined rotational angle based on the flag state, rather than performing complex calculations.
Solution Approach 2:
The detection flag serves as an intermediary that simplifies the decision-making process during rendering. Instead of continuously calculating whether dual-roll compensation is needed, the pre-computed detection flag provides a simple binary indication, allowing the renderer to quickly apply the appropriate rotational angle without complex real-time evaluations, thus reducing processing time while still preventing motion sickness when necessary.
Data Source
AI summary
A method for rendering a current image on a head-mounted display is provided. The current image has been acquired by a camera mounted on a moving object. First, the capturing roll angle is obtained. It is compared to the viewing roll angle that is the current roll angle of the head of a user of the head-mounted display. If the difference between the capturing roll angle and the current viewing roll angle is lower than a determined threshold, a detection flag is activated. The current image is then rotated by an angle determined according to the capturing roll angle, the current viewing roll angle and of the detection flag. The current rotated image is rendered.