G

Grounding Dino Tiny ONNX

Developed by onnx-community
A lightweight zero-shot object detection model in ONNX format, compatible with Transformers.js, suitable for browser-side deployment.
Downloads 98
Release Time : 1/5/2025

Model Overview

This model is the ONNX version of IDEA-Research's Grounding DINO Tiny, specifically designed for zero-shot object detection tasks, capable of detecting objects in images based on text descriptions.

Model Features

Zero-shot Detection
Detects new category objects via text descriptions without requiring training
Browser Compatibility
ONNX format supports direct execution in browser environments via Transformers.js
Lightweight Design
The Tiny version optimizes computational efficiency for edge device deployment

Model Capabilities

Image Object Detection
Text-guided Visual Localization
Multi-object Recognition

Use Cases

Intelligent Content Analysis
Image Content Moderation
Automatically detects specific objects in images (e.g., hazardous items)
Returns object locations and confidence scores
Interactive Applications
Web Image Search
Locates elements in webpage images via natural language queries
Returns bounding box coordinates of matched objects
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase