umuteren/FaceData

Dataset Card for Face Images Dataset

Dataset Summary

The Face Images Dataset is a collection of high-quality face images gathered from sources such as Pexels and Unsplash. The dataset is intended for use in a variety of computer vision tasks, including but not limited to face detection, facial recognition, and facial expression analysis. The dataset includes images of individuals of various ages, genders, and ethnicities, featuring a wide range of facial expressions and poses.

Dataset Structure

The dataset is organized into batches, each containing 10,000 images. There are a total of 14 batches, making up the entire dataset.

Keywords

• Man face
• Woman face
• Child face
• Old face
• Crying face
• Shocked face
• Head shot
• Angry face
• Facial expression
• Beautiful face
• Handsome face
• Boy face
• Baby face
• Girl face
• Human face

Data Instances

Each data instance is an image file with a corresponding file name that serves as a unique identifier. Images vary in resolution and aspect ratio, and they capture various facial attributes and expressions.

Data Fields

• file_name: The unique identifier for each image file, typically in the format image_<unique_id>.jpg.

Data Splits

The dataset is split into a single train set, which consists of 14 batches of images:

• train: 131,655 images (13 batches of 10,000 images each)

Usage

Loading the Dataset

You can load the dataset using the datasets library in Python:

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("umuteren/FaceData", split='train', streaming=True)