Upload image to detect objects
State-of-the-art Object Detection YOLOV9 Demo
Identify car damage in images
Detect objects in images using a web app
Identify the top 3 objects in an image
Identify and label objects in images
Detect objects in an image and identify them
Identify benthic supercategories in images
Find and highlight characters in images
Analyze images to count and classify mosquito species
Detect objects in images and highlight them
Ultralytics YOLOv8 Gradio Application for Testing 🚀
Detect objects in anime images
Transformers.js is a JavaScript library designed for object detection tasks. It allows users to upload an image and detect objects within it using modern transformer-based models. Built for both browser and Node.js environments, Transformers.js provides an efficient and user-friendly way to integrate object detection capabilities into web applications.
• Model Support: Compatible with popular object detection models like YOLO, SSD, and more.
• Real-Time Detection: Processes images quickly for instant results.
• Cross-Platform: Works seamlessly in both browser and server-side (Node.js) environments.
• Ease of Use: Simple API with minimal setup required.
• Open Source: Fully customizable to meet specific project needs.
npm install transformers.js
const {TransformerDetector} = require('transformers.js');
const detector = new TransformerDetector('yolov5s');
const image = document.getElementById('image').files[0];
detector.detect(image).then(results => {
console.log(results);
});
What models are supported by Transformers.js?
Transformers.js supports a variety of pre-trained models, including YOLO, SSD MobileNet, and Faster R-CNN. You can specify the model when initializing the detector.
Can Transformers.js be used in production environments?
Yes, Transformers.js is optimized for both development and production use. It supports server-side processing with Node.js and client-side detection in browsers.
How do I handle non-image inputs?
Transformers.js is designed specifically for image inputs. For non-image data, consider preprocessing or converting the input to an image format before detection.