Virtual Try-On Validation via Measurement Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for procuring wearable devices, such as head-mounted devices, lack accurate customization and fitting capabilities without physical access to a retail establishment, as they often fail to provide precise virtual try-on experiences.
Innovation Solution
A method involving an image capture assembly and simulation module that captures reference images of users wearing wearable devices, detects measurements, generates rendered images, and compares them to determine differences, triggering machine learning model training to adjust the placement of the device for accurate fitting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If existing virtual try-on systems superimpose images of selected eyewear onto images of the user, then the ease of operation is improved, but the measurement precision and manufacturing precision of the wearable device fit deteriorate
Solution Approach 1:
The system captures reference images of the user wearing the actual wearable device, extracts measurements from these images, compares them against the virtual simulation measurements, and uses the differences to train machine learning models. This feedback loop continuously improves the accuracy of virtual try-on measurements, resolving the contradiction between ease of operation and measurement precision.
Solution Approach 2:
The system creates a virtual copy (rendered image) of the wearable device on the user and compares it against the actual reference image. By copying the wearable device into a virtual representation and systematically comparing measurements between the two, the system achieves both ease of virtual operation and high measurement precision through iterative improvement.
2Manufacturing precision
If machine learning models are trained using reference images and measurements, then the manufacturing precision of virtual fitting is improved, but the device complexity and time required for training increase
Solution Approach 1:
The machine learning model is trained using automatically extracted measurements from reference images and rendered images. The system serves itself by automatically capturing images, extracting measurements, comparing them against virtual simulations, and updating the model without requiring manual intervention. This self-service approach improves manufacturing precision while managing device complexity through automation.
3Measurement precision
If the system compares measurements from reference images to rendered images, then the measurement precision is improved, but the loss of time for training and validation increases
Solution Approach 1:
The system performs preliminary actions by capturing reference images and extracting measurements in advance, then uses these pre-processed data to train the machine learning model. By preparing and comparing measurement data before final validation, the system improves measurement precision while reducing the time required for iterative training and validation cycles.
Data Source
AI summary
Systems and methods for validation of modeling and simulation systems that provide for the virtual try-on of wearable devices, such as glasses, by a user, and for the virtual fitting of selected wearable devices for the user. Wearable fit measurements, display fit measurements, ophthalmic fit measurements and other such measurements associated with the fit and function of the wearable device may be detected from image data capturing the wearable device worn by the user. The detected measurements may be compared to corresponding measurements detected in a virtual simulation of the wearable device worn by the user. The comparison may provide for validation and increased accuracy/realism of the modeling and simulation systems.