AIbase
EN
Home
Vit Base Patch32 224 In21k Finetuned Eurosat
V

Vit Base Patch32 224 In21k Finetuned Eurosat

Developed by keithanpai
An image classification model based on Google's Vision Transformer (ViT) architecture, fine-tuned on the EuroSAT dataset for satellite image classification tasks
Image Classification
Transformers
Open Source License:Apache-2.0#High-precision image classification#Remote sensing image recognition#ViT fine-tuned model
Downloads 20
Release Time : 1/13/2023

Model Overview

This model is an image classification model based on the Vision Transformer architecture, fine-tuned on the EuroSAT satellite image dataset, specifically designed for remote sensing image classification tasks.

Model Features

High-precision classification
Achieves 99.45% accuracy on the evaluation set, demonstrating excellent performance
Transformer-based architecture
Utilizes the Vision Transformer architecture instead of traditional CNNs, providing better global feature extraction capabilities
Pre-trained model fine-tuning
Fine-tuned from a ViT model pre-trained on ImageNet-21k, offering robust feature extraction capabilities

Model Capabilities

Satellite image classification
Remote sensing image analysis
Multi-category image recognition

Use Cases

Remote sensing applications
Land use classification
Classifies and identifies different land types in satellite images
99.45% accuracy
Environmental monitoring
Identifies and analyzes changes in land cover
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
EnglishįŽ€äŊ“中文įšéĢ”中文ãĢãģんご
Š 2025AIbase