What is Open-Vocabulary Object Detection?

Open-vocabulary object detection is an advanced AI technology that identifies and segments objects in images using natural language descriptions. Unlike traditional detection models limited to fixed classes, SAM 3 (Segment Anything Model 3) can detect any object you describe — from "person wearing a red jacket" to "coffee cup on desk". Simply describe what you're looking for, and the model will find and segment it.

Supported File Types

• Images: JPEG, PNG, WebP, BMP
• Maximum file size: 10MB

For best results, use clear, high-resolution images with good lighting.

How it works

• Upload an image by dragging and dropping or clicking to browse.
• Describe what you want to detect using natural language (e.g., "person with hat, red car, dog").
• Click "Detect Objects" to process your image.
• View and download the annotated image with segmentation masks.

Privacy & Data Handling

Your uploaded images are processed securely and are not stored permanently. Images are deleted immediately after processing. We do not use your data for training purposes. For more details, see our Privacy Policy.

API Access

Need to integrate object detection into your application? Check out our Object Detection API documentation for programmatic access with additional features.