Projector Calibration With Dual Cameras and Marker-Free Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing projection systems require physical markers on the surface for calibration, which limits flexibility and precision, and often use RGB+IR cameras that are inflexible and less suitable for detecting infrared light effectively.
Innovation Solution
A projector system with two distinct cameras, one sensitive to projected light and one insensitive, allows for precise calibration without markers by mapping reference frames, using filters to isolate detection and projected light wavelengths, and employs separate cameras to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical markers are used on the surface for calibration, then calibration precision is improved, but device complexity and flexibility deteriorate
Solution Approach 1:
The patent extracts the calibration function from physical markers and implements it through software-based reference images projected onto the surface. The system projects a reference image containing known geometric features, captures it with the camera, and computationally determines the mapping between camera and projector coordinate systems, eliminating the need for physical calibration markers.
Solution Approach 2:
The patent uses a digital reference image as a virtual copy of calibration markers. This reference image is projected onto the surface and captured by the camera, creating a digital twin that enables calibration without physical objects. The reference image contains encoded geometric information that allows the system to compute transformation matrices for accurate spatial mapping.
2Adaptability or versatility
If RGB+IR cameras are used to detect both visible and infrared light, then detection capability is improved, but adaptability and detection accuracy deteriorate
Solution Approach 1:
The patent segments the detection function into two separate cameras: one optimized for visible light (RGB camera) and another for infrared light (IR camera). Each camera is dedicated to detecting its specific wavelength range, allowing both to operate at optimal performance without cross-interference, thereby improving both adaptability and detection accuracy.
Solution Approach 2:
The patent introduces wavelength-selective optical filters as intermediaries between the light sources and cameras. These filters ensure that each camera receives only its designated wavelength range by blocking other wavelengths, thereby preventing signal contamination and improving detection precision while maintaining versatile detection capability.
3Measurement precision
If the detection camera is made insensitive to projection light through filtering, then detection accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies wavelength-selective filtering at the local level of each camera sensor. The visible light camera is equipped with an infrared-blocking filter, and the infrared camera is equipped with a visible light-blocking filter. This localized quality assignment allows each camera to be highly sensitive to its target wavelength while being insensitive to other wavelengths, improving detection accuracy without requiring complex system-level modifications.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables flexible and precise interaction with projected images by allowing automatic calibration on any surface without markers, improving detection accuracy and reducing reliance on specialized cameras.
Implementation Method 1
a first camera (6) for recording an image of the surface (10), wherein the first camera (6) is sensitive to light from the projector (4) that has reflected from the surface (10)
Implementation Method 2
a second camera (8) for recording images of the surface (10), wherein the second camera (8) is sensitive to infrared light from objects on or near the surface (10)
Implementation Method 3
The detection camera is insensitive to the projection light (in particular also during the calibration stage), known calibration methods based on detection of a projected image by the detection camera may not be used
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A projector system (2) is disclosed. The projector system (2) comprises a projector (4) for projecting light on a surface (10). The projected light is configured to form one or more images on the surface (10). The system (2) also comprises a first camera (6) for recording an image of the surface (10). The projector system (2) also comprises a second camera (8) for recording images of the surface (10). The second camera (8) is sensitive to detection light from an object near and/or on the surface (10), and insensitive to the projected light. The projector system (2) also comprises a data processing system (100) configured to perform a number of steps. One step that the data processing is configured to perform is providing calibration image data to the projector (4). The data processing system (100) then causes the projector (4) to project light on the surface (10) to form a projected version of the first calibration image on the surface (10). Another step that the data processing system (100) is configured to perform is a step of receiving, from the first camera (6), image data representing a recorded image of the surface (10) as recorded by the first camera (6) while the first calibration image was being projected on the surface (10). Then, based on the image data, the data processing system (100) can determine the position of objects, detected by the second camera (8), in the projector reference frame.