Detect and label objects in photos with bounding boxes. 80 object categories.
Upload an image to detect objects
Three steps. Instant results.
This tool uses TensorFlow.js with the pre-trained COCO-SSD (Common Objects in Context - Single Shot MultiBox Detector) model to identify up to 80 different object categories in your image. It processes the image pixel data directly in your browser, running the neural network to output bounding boxes, labels, and confidence scores, which are then drawn onto a canvas overlay.