AI-Powered Image Object Detection and Bounding Box Visualization

This n8n workflow demonstrates how to perform prompt-based object detection on an image using Google’s Gemini 2.0 API and visualize the detected objects with bounding boxes. The workflow starts with a manual trigger for testing purposes and involves several steps: downloading an image, retrieving its dimensions, sending the image to Gemini for object detection, normalizing and scaling the bounding box coordinates to fit the original image, and finally drawing these bounding boxes onto the image. It includes informative sticky notes explaining each step, making it suitable for automating tasks like identifying objects in images for content moderation, digital asset management, or visual data analysis. This setup enables users to leverage AI to automate detailed image analysis and annotation processes efficiently.