A new DIY project called the Poetry Camera combines photography with artificial intelligence to create poems from captured images. The device, developed by Kelin ‘Caroline’ Zhang and Ryan Mather, uses computer vision to identify elements in photographs and transform them into verse rather than producing traditional prints.
The camera’s hardware consists of a Raspberry Pi Zero 2 chip, Camera Module 3, and an Adafruit Mini Thermal Printer. The system connects to OpenAI’s API to process images and generate poetry, all housed within a 3D-printed enclosure. Users can control the device through a shutter button and an adjustment knob that determines the length of the generated poems.
The technical specifications and build instructions are publicly available on GitHub for DIY enthusiasts. The device features a minimalist design with just two main controls, making it accessible to users without extensive technical knowledge. The creators have opted to make the project open-source, allowing others to replicate and modify the design.
While mass production isn’t currently planned, the developers intend to offer limited quantities through their website. Pricing details have not yet been announced. The project joins other AI-photography experiments, including Dan Macnish’s 2018 doodling camera and the 2023 Paragraphica, which generates AI images from location and time data.
“I want AI to do my laundry and dishes so that I can do art and writing,” project inspiration Joanna Maciejewska is quoted as saying, though the Poetry Camera takes a different approach by integrating AI directly into the creative process. The device represents a growing trend of combining artificial intelligence with traditional photography tools to create new forms of artistic expression.
Source: yankodesign.com