3D Model Scaling via Interpupillary Distance Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately scale 3D models of users and objects for virtual try-on applications, such as fitting glasses, due to limitations in measuring interpupillary distance and pixel density, leading to inaccurate depictions.
Innovation Solution
A computer-implemented method and system that uses a scaling module to determine the distance of an object from a camera, query pixel densities, and scale 3D models based on these measurements, ensuring accurate scaling and alignment of virtual try-on models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct measurement units (laser or infra-red pointer) are used to measure distance between camera and person, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent uses an intermediary approach by measuring the distance between the camera and the person's face using the person's known physical dimensions (interpupillary distance) as a reference standard. Instead of directly measuring the camera-to-face distance with complex laser/infra-red equipment, the system captures an image, detects pupil positions, and calculates the distance based on the known real-world distance between pupils and the pixel distance in the image. This mediator approach (using anatomical reference) simplifies the measurement system while maintaining precision.
2Device complexity
If the distance between camera and person is not accurately measured, then device complexity is reduced, but manufacturing precision of virtual try-on model deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the system calculates the camera-to-face distance based on the detected pupil positions and the known interpupillary distance. This calculated distance is then fed back into the scaling calculation to determine the correct scale factor for the virtual try-on model. The feedback loop ensures that the scaling is accurate by continuously referencing the measured pixel distance against the known real-world distance, thereby maintaining manufacturing precision without requiring complex external measurement devices.
3Ease of operation
If scaling factor is calculated using assumed average interpupillary distance (7 cm), then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent enables the system to self-determine the user's specific interpupillary distance by detecting the pupil positions in the captured image and calculating the pixel distance between them. Instead of relying on a pre-assumed average value, the system performs its own measurement for each user, automatically adapting to individual variations. This self-service approach maintains ease of operation (no manual input needed) while significantly improving measurement precision by using actual user-specific data rather than population averages.
4Device complexity
If pixel density is not accurately determined, then device complexity is reduced, but reliability of virtual try-on application deteriorates
Solution Approach 1:
The patent performs preliminary determination of pixel density by capturing a reference image and calculating the pixel distance between pupils before proceeding with the virtual try-on scaling. By establishing the pixel-to-real-world-distance ratio in advance (using the known interpupillary distance and measured pixel distance), the system prepares the scaling factor beforehand. This preliminary action ensures reliable and accurate scaling in the subsequent virtual try-on process without requiring complex real-time pixel density measurements during the actual try-on operation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method for scaling an object is described. A distance of an object from the computing device is measured via a processor of a computing device in conjunction with a rangefinder. An image of the object is captured via the processor. A database of pixel densities is queried, via the processor, for a pixel density at the measured distance of the object from the computing device. A depiction of the object is scaled based on determining a distance of a detected feature of the object.