AIbase
EN
Home
Beit Sketch Classifier Pt Metaset
B

Beit Sketch Classifier Pt Metaset

Developed by kmewhort
A fine-tuned sketch classification model based on microsoft/beit-base-patch16-224-pt22k-ft22k, achieving 82.77% accuracy on the evaluation set
Image Classification
Transformers
Open Source License:Apache-2.0#Sketch Classification#High Accuracy#Transfer Learning
Downloads 18
Release Time : 1/10/2023

Model Overview

This model is a vision classification model based on the BEiT architecture, specifically fine-tuned for sketch recognition tasks, suitable for sketch classification and identification

Model Features

High Accuracy
Achieves 82.77% classification accuracy on the evaluation set
Based on BEiT Architecture
Utilizes the advanced BEiT Vision Transformer architecture with powerful feature extraction capabilities
Fine-tuned Optimization
Precisely fine-tuned on specific datasets, making it suitable for sketch classification tasks

Model Capabilities

Sketch Classification
Image Recognition
Visual Feature Extraction

Use Cases

Creative Design
Automatic Sketch Classification
Automatically classifies and labels sketches drawn by designers
Accuracy reaches 82.77%
Educational Applications
Drawing Teaching Assistance
Identifies sketches drawn by students and provides classification feedback
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