Inspiration

The inspiration for this project is a desire to make fashion accessible and fun for everyone. With this lens, users can try on different outfits and experiment with new styles without having to physically buy or own the clothing items.

What it does

The lens is a Snapchat lens that allows users to copy the upper garment (e.g. shirt, blouse, jacket, etc.) of any person they see and virtually try it on themselves. Using computer vision and augmented reality, the lens analyzes the image and superimposes the selected outfit onto the user's body in real-time. This provides a fun and interactive way for users to experiment with different fashion styles and find inspiration for their own outfits.

How we built it

Image Processing: The first step in building was developing algorithms to detect and extract the clothing from the image(I used Lens Studio Garment Transfer Feature).

Machine Learning Models: Next, machine learning models is trained to recognize and classify different types of clothing items(I used Lens Studio Garment Transfer Feature).

AR Integration: Once the clothing has been identified and extracted, augmented reality tools has been used to integrate the virtual outfit onto the user's body in real-time. This involved creating 3D models of the clothing and mapping it onto the user's body using AR technology.

User Experience: Finally, user testing and iteration was done to refine the experience and ensure that it is seamless and easy to use. This includes making adjustments to the user interface, improving performance, and fixing any bugs.

Challenges we ran into

The main challenges faced while building the lens was the technical aspects of image processing and AR, as well as ensuring a smooth and seamless user experience. Additionally, making sure the lens performs well on a variety of devices and under different lighting conditions

Accomplishments that we're proud of

Innovative use of technology: By combining image processing, machine learning, and AR, I have created a unique and innovative lens that provides a new way for users to experiment with fashion.

What we learned

how to extract information from images and use it to recognize and classify different types of clothing.

What's next for Garment Transfer

The lens could be expanded to include other types of clothing items, such as pants, skirts, or shoes. This would provide users with even more ways to experiment with fashion and try on different outfits.

Built With

Share this project:

Updates