AIbase
EN
Home
My Food Model
M

My Food Model

Developed by iammartian0
Food image classification model based on Google Vision Transformer (ViT) architecture, fine-tuned on the Food101 dataset with an accuracy of 90.9%
Image Classification
Transformers
Open Source License:Apache-2.0#Food Image Classification#High-precision Recognition#ViT Fine-tuning
Downloads 18
Release Time : 2/22/2023

Model Overview

This model is a food image classification model obtained by fine-tuning Google's ViT-base-patch16-224-in21k pre-trained model on the Food101 dataset. It is mainly used to recognize images of 101 different categories of food.

Model Features

High Accuracy
Achieves 90.9% classification accuracy on the Food101 dataset
Based on ViT Architecture
Uses the Vision Transformer architecture with powerful image feature extraction capabilities
Few-shot Fine-tuning
Achieves good results with fine-tuning on just 5,000 training samples

Model Capabilities

Food Image Classification
101-class Food Recognition

Use Cases

Food Recognition
Restaurant Dish Recognition
Used in restaurants to automatically identify photos of dishes taken by customers
Accurately recognizes 101 common food items
Nutrition Analysis Application
Automatically identifies food photos taken by users in health applications
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