AIbase
EN
Home
Beit Base Patch16 224 Pt22k Ft22k Finetuned FER2013 7e 05
B

Beit Base Patch16 224 Pt22k Ft22k Finetuned FER2013 7e 05

Developed by lixiqi
A facial expression recognition model fine-tuned on the FER2013 dataset based on Microsoft's BEiT model
Image Classification
Transformers
Open Source License:Apache-2.0#Fine-tuned Image Classification#High Accuracy#FER2013 Dataset
Downloads 19
Release Time : 1/9/2023

Model Overview

This model is a vision Transformer based on the BEiT architecture, specifically fine-tuned for facial expression recognition tasks, achieving an accuracy of 72.2% on the FER2013 dataset.

Model Features

High-Precision Expression Recognition
Achieves 72.2% accuracy on the FER2013 dataset, effectively recognizing various facial expressions
Based on BEiT Architecture
Utilizes an advanced vision Transformer architecture, combining image patch embedding and self-attention mechanisms
Fine-Tuned Optimization
Fine-tuned with a learning rate of 7e-05 to optimize performance in expression recognition tasks

Model Capabilities

Facial Expression Recognition
Image Classification
Sentiment Analysis

Use Cases

Affective Computing
Real-Time Expression Analysis System
Used for real-time expression recognition in video conferences or social media
Can recognize 7 basic expressions with 72.2% accuracy
Human-Computer Interaction
Emotion-Aware Robot Interaction
Enables robots to adjust interaction methods based on user expressions
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