AR Interaction Apparatus Using Reference Image Comparison for Hand Occlusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing calibration process for interactions between real objects and fingers in augmented reality and IoT environments is burdensome and affects usability, as it requires repeated calibration when objects are installed or moved.
Innovation Solution
An information processing apparatus that uses a processor and memory to generate and update target region images, detecting object movements by comparing range images with and without hand regions, allowing for automatic complementation of hand-obscured areas and eliminating the need for manual recalibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a calibration process is performed to register the position of the target object, then the interaction accuracy between the real object and finger is improved, but the time required for setup and reconfiguration increases
Solution Approach 1:
The system performs preliminary actions by capturing a target region image of the entire shooting range before hand operations begin. This preliminary image serves as a reference that eliminates the need for repeated calibration when the hand enters or leaves the shooting range, as the system can automatically determine whether the target object has moved by comparing current images with the pre-captured reference image.
Solution Approach 2:
The system creates a copy of the target region image (a image indicating the entire shooting range without hand obstruction) and stores it as reference data. This copied image is then used for comparison with current range images to detect target object movement, eliminating the need for manual recalibration and reducing setup time while maintaining interaction accuracy.
2Measurement precision
If the calibration process is performed every time the target object is installed or moved, then the position registration accuracy is maintained, but the usability and operational convenience deteriorate
Solution Approach 1:
The system performs self-service by automatically detecting whether the target object has moved through image comparison algorithms. The processor automatically compares the current range image with the stored target region image, determines if movement has occurred, and updates the a image accordingly. This self-service capability eliminates the need for manual calibration operations, significantly improving usability while maintaining position registration accuracy.
Solution Approach 2:
The system changes the operational parameter from manual calibration to automatic image-based detection. By transitioning from a manual calibration process to an automatic comparison process using range images, the system maintains position registration accuracy while dramatically improving ease of operation. The processor automatically adjusts the a image based on detected movements without requiring user intervention.
3Duration of action of moving object
If the range image includes the hand region, then the continuous capture of target object position is improved, but the accuracy of target object position detection deteriorates due to hand obstruction
Solution Approach 1:
The system extracts and removes the hand region from the range image through image processing. By identifying and excluding the hand region from the analysis, the system maintains continuous capture capability while ensuring that position detection accuracy is not compromised by hand obstruction. The processor specifically processes the a image to exclude hand regions, allowing continuous monitoring without accuracy loss.
Solution Approach 2:
The system segments the range image into different regions: the hand region and the target object region. By dividing the image into these distinct segments, the system can continuously capture hand movements for interaction detection while simultaneously maintaining accurate target object position detection by analyzing only the non-hand portions of the image.
Data Source
AI summary
An information processing apparatus includes: a memory that stores a target region image generated in the past that indicates a distance to a target object; and a processor coupled to the memory and configured to calculate a hand region within a range image obtained after the target region image has been generated, the range image indicating a distance to a hand and the target object, detect a movement of the target object, when a movement of the target object is not detected, generate a complemented image by complementing, using the target region image, the portion that corresponds to the hand region within the image from which the hand region has been deleted, update the target region image with the complemented image, and, when a movement of the target object has been detected, update the target region image with the image from which the hand region has been deleted.