DINOv2 Open-source Image Feature Extraction Model - Empowered by Register Mechanism for Efficient Image Feature Extraction

Dinov2.large.patch 14.reg 4

Developed by refiners

DINOv2 is a vision transformer-based image feature extraction model that enhances feature extraction capabilities through the introduction of register mechanisms.

Safetensors

Open Source License:Apache-2.0 #Vision Transformer Registers #High-Precision Feature Extraction #Self-Supervised Learning

Downloads 15

Release Time : 8/7/2024

Model Overview

This model is an improved version of DINOv2, primarily used for image feature extraction tasks, with enhanced performance of the base model through an adapter architecture.

Model Features

Improves the feature extraction capability of vision transformers through the introduction of registers.

Adapter Architecture

Adapter implementation based on the facebook/dinov2-large base model.

Efficient Feature Extraction

Model architecture optimized specifically for image feature extraction tasks.

Model Capabilities

Image Feature Extraction

Visual Representation Learning

Use Cases

Computer Vision

Image Feature Extraction

Extracts high-quality feature representations from images.

Can be used for downstream vision tasks such as classification, detection, etc.

Property	Details
Model Type	Adapter for facebook/dinov2-large
Pipeline Tag	Image Feature Extraction
Base Model	facebook/dinov2-large
Base Model Relation	Adapter
Library Name	refiners
Tags	dino, dinov2, features, facebook

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Dinov2.large.patch 14.reg 4

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 DINOv2 large with registers

🚀 Quick Start

📄 License

📚 Documentation

Model Information

Video Demo

Citation