Projection Mapping on Non-Rigid Surfaces Using Flexibility-Segmented Markers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional projection mapping technologies are inadequate for projecting images onto non-rigid objects, such as the human body, due to the challenge of aligning and maintaining image accuracy with changing body shapes, particularly in areas with significant flexibility like breasts, where conventional methods designed for rigid objects fail to account for deformation.
Innovation Solution
An image processing device and projection system that uses distinct markers on non-varying and varying sites on the body, employing a combination of alignment and deformation processing to generate a projection image that accounts for the body's shape and flexibility, ensuring accurate alignment and projection even on non-rigid surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional projection mapping technology is used on rigid objects, then alignment accuracy is maintained, but the technology fails when applied to non-rigid objects with changing shapes
Solution Approach 1:
The patent divides the projection target into multiple regions with different flexibility characteristics (first region with low flexibility and second region with high flexibility). This segmentation allows the system to apply different processing strategies to different body parts, maintaining alignment accuracy while adapting to non-rigid surfaces.
Solution Approach 2:
The patent applies different marker types and alignment methods to different regions based on their flexibility characteristics. First type markers are used on low-flexibility regions while second type markers are used on high-flexibility regions, allowing local optimization of alignment accuracy for each body part.
2Adaptability or versatility
If markers are placed on flexible body parts, then coverage of varying sites is achieved, but shape changes cause misalignment between capture and projection
Solution Approach 1:
The patent introduces a deformation processor that dynamically adjusts the projection image based on the detected shape changes of the second region. The system calculates deformation amounts from marker positions and applies real-time corrections to maintain alignment precision even as the body shape changes between capture and projection.
Solution Approach 2:
The patent uses captured images of markers on the body surface to provide feedback about actual shape changes. This feedback is processed to calculate deformation amounts, which are then used to correct the projection image, creating a closed-loop system that maintains alignment precision despite body movement.
3Ease of operation
If a single alignment method is used for all body parts, then processing simplicity is maintained, but accuracy deteriorates on flexible sites
Solution Approach 1:
The patent segments the alignment processing into two distinct pathways: one for low-flexibility regions using standard alignment methods, and another for high-flexibility regions using deformation compensation. This segmentation maintains operational simplicity for rigid areas while improving precision for flexible areas.
Solution Approach 2:
The patent applies different alignment processing qualities to different body regions based on their flexibility. Simple alignment is sufficient for rigid areas, while deformation-compensated alignment is applied to flexible areas, optimizing the balance between processing simplicity and alignment precision locally.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
An image processing device extracts (40, 40a), for a target area that has sites having different flexibilities and where markers (M1, M2, M3) corresponding to the respective flexibilities are put on the sites, marking positions where the markers (M1, M2, M3) are put according to each type of the markers (M1, M2, M3) from each of a first image to be projected and being an image of the target area captured in advance and a second image being an image of the target area captured before projection of the first image; executes, based on the extracted marking positions according to each type of the markers (M1, M2, M3), image processing to align the marking positions of the markers (M1, M2, M3) contained in the first image with the marking positions of the markers (M1, M2, M3) contained in the second image by a processing method corresponding to the types of the markers (M1, M2, M3); and generates, based on the processed first image, a projection image to be projected onto the target area.