DIY enthusiasts have made an AI camera that, instead of displaying images, reads the images and gives a description about them. For doing so, it sends the image to an AI, which then generates a text description of the scene. The description is displayed on a small OLED screen, giving users a fresh and creative way to "see" the world through AI’s perspective. The camera comes with multiple modes, a few of them being solving math problems or describing the weather.
This AI vision camera is built around an ESP32-CAM module, the setup also includes an OLED screen, a buzzer for notifications, and a few buttons for taking pictures and switching modes. Power is supplied via USB from a portable battery bank, making it compact and easy to use.
OpenAI Chat completion API is used to prompt and analyze the image, which supports prompting with an image. The AI processes the request based on prompts like “describe the weather,” allowing it to generate relevant and useful descriptions.
This DIY project is a budget-friendly way to experiment with AI, computer vision, and API requests. Whether you're a maker, an AI enthusiast, or just someone looking for a fun weekend project, this AI-powered camera is a beginner-friendly project to work on.
