Dataset Card for Food-101

  Dataset Summary

This dataset consists of 101 food categories, with 101'000 images. For each class, 250 manually reviewed test images are provided as well as 750 training images. On purpose, the training images were not cleaned, and thus still contain some amount of noise. This comes mostly in the form of intense colors and sometimes wrong labels. All images were rescaled to have a maximum side length of 512 pixels.

  Supported Tasks and… See the full description on the dataset page: https://huggingface.co/datasets/ethz/food101.

Food-101 dataset is a large-scale dataset for food image classification, with 101 classes and over 30,000 images.

The Food-101 is a challenging data set of 101 food categories with 101,000 images. For each class, 250 manually reviewed test images are provided as well as 750 training images. On purpose, the training images were not cleaned, and thus still contain some amount of noise. This comes mostly in the form of intense colors and sometimes wrong labels. All images were rescaled to have a maximum side length of 512 pixels.

1) Data Introduction ? The Food-101 dataset contains subsets of the original Food-101 data, featuring multiple food categories and intended to serve as a richer alternative to classic image datasets like CIFAR-10 or MNIST.

2) Data Utilization (1) Characteristics of the Food-101 Dataset: ? The dataset consists of 49 food categories, with data files indicating the number of images and their respective resolutions. ? Includes both color (RGB) and grayscale images with labels.

(2) Applications of the Food-101 Dataset: ? Food image classification: Useful for developing and evaluating models that can automatically recognize and classify various food items. ? Model interpretability and explainability: Can be used to study which regions or components of food images are most important for classification decisions. ? Advanced food analysis: Provides opportunities to identify new food types as combinations of existing tags or to build detectors for food items in complex scenes.

Dataset

This dataset was created by JAYAPRAKASHPONDY

Contents

Food 101 Classification 2

## Overview

Food 101 Classification 2 is a dataset for classification tasks - it contains Objects KcoQ annotations for 14,995 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Food-101: mining discriminative components with random forests.

🍽️ Food101-RecipeDataset

The Food101-RecipeDataset is a curated collection of food images paired with detailed recipe metadata — including dish names and ingredient breakdowns. It is designed to support machine learning tasks in computer vision, recipe generation, nutrition analysis, and more.

  🗂️ Dataset Overview

Each example in this dataset includes:

Image: A high-quality food image. Food Name: The name of the dish (e.g., "Spaghetti Carbonara"). Ingredients: A list… See the full description on the dataset page: https://huggingface.co/datasets/Moiz2517/Food-101-RecipeDataset.

Food 101 Detection

## Overview

Food 101 Detection is a dataset for classification tasks - it contains Food annotations for 9,617 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Context

There's a story behind every dataset and here's your opportunity to share yours.

Content

Jordan cuisine consists of a variety of regional and traditional to the Arabic subcontinent. Given the diversity in soil, climate, culture, ethnic groups, and occupations, these cuisines vary substantially and use locally available spices, herbs, vegetables, and fruits.

This dataset consists of information about various Jordan dishes, their ingredients, their place of origin, etc.

Challenge

The first goal is to be able to automatically classify an unknown image using the dataset, but beyond this there are a number of possibilities for looking at what regions/image components are important for making classifications, identify new types of food as combinations of existing tags, build object detectors which can find similar objects in a full scene.

About

For more information: http://arartawil.com/

Dataset

This dataset was created by Steven Chen

Released under Apache 2.0

Contents

Food image dataset used for food image recognition

kaze-desu/food-101-4shot-b2n dataset hosted on Hugging Face and contributed by the HF Datasets community

The subset of the Food-101 dataset comprises a curated selection of 20 distinct food categories, providing a focused yet diverse collection for analysis and exploration. Each category represents a popular culinary class, ranging from appetizing desserts like ice cream and pancakes to savory dishes such as pizza and sushi. With a reduced size compared to the comprehensive Food-101 dataset, this subset streamlines data handling and computational resources while retaining the richness and variety essential for meaningful analysis. It contains 17500 images for training and 2500 images for testing and validation (1250 each). | Sl No. | Category |Count | | --- | --- | | 1 | Fried Rice | 750 | | 2 | Dumplings | 750 | | 3 | French Fries | 750 | | 4 |Ramen | 750 | | 5 |Nachos | 750 | | 6 |Pizza | 750 | | 7 |Hamburger | 750 | | 8 |Chocolate Cake | 750 | | 9 |Omelette | 750 | | 10 | Cannoli | 750 | | 11 | Apple Pie | 750 | | 12 | Hot Dog | 750 | | 13 | Spring roll | 750 | | 14 | Soup | 750 | | 15 | Waffle | 750 | | 16 | Chicken curry | 750 | | 17 | Donuts | 750 | | 18 | Samosa | 750 | | 19 | Ice Cream | 750 | | 20 |Cup Cake | 750 |

Dataset Details

  Dataset Description

This dataset is the combination of the Food101, Indian Food Classification and The-massive-Indian-Food-Dataset datasets. This Dataset aims to be a viable dataset for Image Classification of Foods with an added Indian context. This dataset has 121 classes with each class having 800 images in the train split and 200 images in the test split. Maximum resolution of images is 512*512. The Food121-224 dataset has all images downscaled to a… See the full description on the dataset page: https://huggingface.co/datasets/ItsNotRohit/Food121.

Food101

## Overview

Food101 is a dataset for object detection tasks - it contains Object annotations for 1,995 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Food-101 (Test set only)

Original paper: Food-101 – Mining Discriminative Components with Random Forests Homepage: https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/ Bibtex: @inproceedings{bossard14, title = {Food-101 -- Mining Discriminative Components with Random Forests}, author = {Bossard, Lukas and Guillaumin, Matthieu and Van Gool, Luc}, booktitle = {European Conference on Computer Vision}, year = {2014} }

Dataset

This dataset was created by K. T.

Released under GPL 2

Contents

📝 Dataset Overview This dataset provides nutritional information for the 101 food categories in the popular Food-101 dataset. Each food type is enriched with calories, protein, fat, carbs, and other nutritional data per 100 grams or per serving. It is intended to help data scientists, researchers, and developers who are working on food classification, nutrition estimation, or AI-powered health applications.

🔗 Source and Inspiration The Food-101 dataset, a widely used benchmark dataset for food classification, lacked comprehensive nutritional data. This dataset was created to bridge that gap by providing accurate, research-based nutritional values for each food class. The information has been compiled from reliable sources such as USDA Food Database, and verified nutrition resources.

📊 What’s Included 101 Food Categories matching the classes in Food-101.

Nutritional Information for each category including:

• Calories
• Protein (g)
• Fat (g)
• Carbohydrates (g)
• Fiber (g)
• Sugar (g)
• Sodium (mg)

This dataset can be used for:

Calorie estimation models. - Food recommendation systems. - Diet tracking applications. - Enhancing Food-101 classification models with nutritional insights.

✅ Potential Use Cases - Combine with Food-101 images to build a calorie estimation app. - Use for nutrition-aware meal planning apps. - Research in food classification + nutrition prediction pipelines.

🍽️ Extended Food-270 Dataset

This dataset is an extended version of the original Food-101 dataset, designed to improve the diversity and representativeness of food image classification models.It combines the well-known Food-101 dataset with additional custom food images collected and curated by Berker Üveyik to expand the number of classes and enhance the dataset’s real-world applicability.

  📚 Overview

Base Dataset: Food-101
Extended Version: Food-270
Total… See the full description on the dataset page: https://huggingface.co/datasets/berkeruveyik/food_270.