Detect objects in images using Transformers.js
Find license plates in images
Detect objects in images using drag-and-drop
Detect objects in anime images
Identify objects in images with YOLOS model
Detect objects in an uploaded image
Track objects in live stream or uploaded videos
Cutting edge open-vocabulary object detection app
Identify labels in an image with a score threshold
Identify the top 3 objects in an image
Identify objects in images
Detect objects in your images
Identify segments in an image using a Detectron2 model
Transformers.js is a JavaScript library designed for object detection tasks. It allows developers to integrate transformer-based models into web applications, enabling the detection of objects within images efficiently.
• Model Integration: Supports popular transformer-based models for accurate object detection.
• Browser Compatibility: Works seamlessly in modern web browsers.
• Real-time Processing: Enables real-time object detection in images.
• Easy API: Provides a simple and intuitive API for developers.
• Customizable: Allows for fine-tuning and customization of detection settings.
npm install transformers.js
import Transformers from 'transformers.js';
const detector = new Transformers.Detector();
detector.detectObjects(imageElement)
.then(results => console.log(results));
What browsers are supported by Transformers.js?
Transformers.js is designed to work with modern browsers that support WebGL and TensorFlow.js. Chrome, Firefox, and Edge are recommended.
How accurate is Transformers.js for object detection?
Accuracy depends on the underlying model used. Popular models like DETR and YOLO achieve high precision, but results may vary based on image quality and complexity.
Can I use custom models with Transformers.js?
Yes, Transformers.js supports custom models. You can load your own-trained models using the loadModel()
method and specify the architecture type.