Video See-Through HMD Single Sensor Parallax Elimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video see-through type head-mounted displays (HMDs) face challenges in accurately estimating the position of hands due to parallax between images captured by different cameras, leading to unintended positional deviations and user discomfort.
Innovation Solution
The use of a single image pickup element to capture both display and detection images eliminates parallax, allowing for parallel imaging and processing that reduces delays and improves accuracy in detecting real objects, such as hands, without requiring complex arithmetic operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If multiple cameras are used to capture images for display and detection, then detection capability is improved, but parallax between images causes positional deviation and user discomfort
Solution Approach 1:
The patent combines the display image capture and detection object capture functions into a single image pickup element. This merging eliminates the parallax issue that occurs when separate cameras are used, as both the display image and detection image are captured from the same optical viewpoint, ensuring accurate position detection without positional deviation.
2Difficulty of detecting and measuring
If separate cameras are used for display and detection, then detection function is enhanced, but complex arithmetic operations are required to correct parallax
Solution Approach 1:
By using a single image pickup element to capture both display and detection images simultaneously, the patent eliminates the need for complex parallax correction arithmetic operations. The detection image is directly obtained from the same optical path as the display image, simplifying the processing requirements while maintaining enhanced detection functionality.
3Productivity
If parallel imaging is implemented with single sensor, then processing speed is improved, but image quality may be compromised
Solution Approach 1:
The patent segments the image data from the single image pickup element into two separate images: a display image and a detection image. This segmentation allows parallel processing of both images simultaneously, improving processing speed while maintaining image quality, as each image can be optimized for its specific purpose without compromising the other.
Data Source
AI summary
A head mounted display according to the present invention is a head mounted display of a video see-through type, and includes: an image sensor configured to be able to execute, in parallel, capturing of a first image representing a real space, and capturing of a second image representing the real space; a display configured to display the first image; a detection unit configured to detect a predetermined real object from the second image; and a processor configured to perform predetermined processing based on a result of the detection of the predetermined real object.