CaptureCube

Inspiration

We believe that everyone deserves the chance to showcase their work, especially creators looking to sell their products and hardware online. That's why we developed CaptureCube, a cohesive system that handles everything from photographing your product to creating an advertisement.

What it does

CaptureCube takes multiple pictures of your product via a dual-axis display machine to capture virtually every face and feature of the product. The AI toolsets we leverage are able to tranform these simple images into a realistic 3D model that can be automatically integrated into advertisments and web page displays.

How we built it

A Raspberry Pi 4 controls the motors spinning the platform, and triggers a USB camera module to capture images. The data is then saved to the cloud with an AWS S3 Bucket. The images are then autonomously downloaded and fed into Gemini Veo, which is able to create customized advertisements and other video clips relative to the object.

Challenges we ran into

On the hardware side it was a challenge to get all the angles we needed to make an accurate reconstruction of the product. We overcame this with a simple and robust custom-designed product mount that would rotate in two axes, giving the camera a range of views. On the software side we had to learn to control motors and the camera with a python script on the Raspberry Pi, and then how to interface with AWS through the Pi.

Accomplishments that we're proud of

We managed to create something that effectively drew on hardware as well as software innovation to make something that small business owners and product creators could truly want. We are also proud of the fact that we learned to integrate Gemini API seamlessly with our backend as well as our front end. This was truly a full stack developer and it was exciting to see it come together.