Overview / Usage

What if a Polaroid camera could transform your environment into a work of art in realtime? What if every detail of the world would be shown in the style of a painting from Van Gogh? What if you could instantly take home a Polaroid-like work of art?

One of the many interesting developments in the field of Deep Learning is the so called "Style Transfer". It describes a possibility to create a patchwork (or pastiche) from two images. While one of these images defines the artistic style of the result picture, the other one is used for extracting the image content. The Hardware Hacking Team from TNG Technology Consulting managed to build an application using OpenCV and Tensorflow to realize such a camera. The resulting images can be printed as Polaroid images.

Methodology / Approach

The world is recorded by a camera located in the 3D-printed housing of the Polaroid camera. The recording Raspberry Pi mini computer is then sending the images from this camera to a powerful workstation computer. There the images are stylized using a convolutional neural net and Tensorflow. The stylized content is streamed back to the Polaroid camera, where it is shown on a display. This image can then be printed instantly using a mobile photo printer.

Technologies Used

• Tensorflow
• OpenCV and GStreamer
• Raspberry Pi 3/4, Raspberry Pi display and camera
• Custom 3D printed camera housing

Collaborators