AIbase
EN
Home
Beit Base Patch16 224 Pt22k Ft22k Finetuned FER2013CKPlus 7e 05 Finetuned FER2013 7e 05
B

Beit Base Patch16 224 Pt22k Ft22k Finetuned FER2013CKPlus 7e 05 Finetuned FER2013 7e 05

Developed by Celal11
An image classification model based on the BEiT architecture, fine-tuned on the FER2013 and CK+ datasets, primarily used for facial expression recognition tasks.
Image Classification
Transformers
Open Source License:Apache-2.0#Facial Expression Recognition#High-precision Classification#Fine-tuned Model
Downloads 17
Release Time : 1/27/2023

Model Overview

This model is a vision Transformer based on the BEiT architecture, specifically designed for image classification tasks, excelling particularly in the field of facial expression recognition.

Model Features

Efficient Fine-tuning
The model has been meticulously fine-tuned on the FER2013 and CK+ datasets, optimizing its facial expression recognition capabilities.
Transformer Architecture
Utilizes the BEiT vision Transformer architecture, effectively capturing both global and local features in images.
Stable Training
Employs linear learning rate scheduling and warm-up strategies to ensure training stability.

Model Capabilities

Image Classification
Facial Expression Recognition
Visual Feature Extraction

Use Cases

Affective Computing
Facial Expression Analysis
Used to identify facial expressions in images, such as happiness, sadness, anger, etc.
Achieved 71.16% accuracy on the FER2013 dataset.
Human-Computer Interaction
Emotion-aware Systems
Can be used to build interactive systems capable of sensing user emotional states.
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