metadata

language:
  - en
license: openrail
size_categories:
  - 10K<n<100K
tags:
  - art
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': Adire
            '1': Idama
            '2': Idgo
            '3': Idoma
            '4': Igala
            '5': Igbo
            '6': Tiv
            '7': Tswana-Shweshwe
            '8': Xhosa-South Africa
            '9': Zulu
  splits:
    - name: train
      num_bytes: 192467436.519
      num_examples: 12363
  download_size: 862802610
  dataset_size: 192467436.519

African Attire Images Dataset

Overview

The African Attire Images Dataset is a curated collection of images showcasing various African traditional attires and clothing styles. This dataset is intended to promote cultural diversity and provide valuable resources for research, education, and creative projects related to fashion, culture, and identity.

Dataset Details

Total Images: XXXX
Categories(ClassLabel): 'Adire', 'Idama', 'Idgo', 'Idoma', 'Igala', 'Igbo', 'Tiv', 'Tswana-Shweshwe', 'Xhosa-South Africa', 'Zulu'
Resolution: High-resolution images, ensuring detailed representation.
Format: JPEG, PNG, etc.
Source: Curated from diverse online sources.

Why This Dataset?

Cultural Heritage: Explore the rich and diverse cultural heritage of Africa through its clothing and adornments.
Research: Ideal for studies related to fashion trends, cultural anthropology, and identity representation.
Creative Projects: Designers, artists, and creators can use this dataset for inspiration and reference in their work.
Educational Use: Educators can incorporate these images into lesson plans to teach about cultural diversity and traditional clothing.

Data Preprocessing

The dataset has undergone preprocessing to ensure quality and consistency but more cleaning can still be done:

Image Quality: Only high-quality images have been selected, ensuring clear and detailed representations.
Metadata: Each image is accompanied by metadata, including category labels, image source, and optional additional information.
Data Augmentation: In some cases, data augmentation techniques have been applied to increase dataset diversity.

Dataset Structure

The dataset is organized as follows:

african_attire_dataset/
│
├── Adire/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Idama/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Idgo/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Idoma/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Igala/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Igbo/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Tiv/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Tswana-Shweshwe/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Xhosa-South Africa/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── Zulu/
│   ├── image1.jpg
│   ├── image2.jpg
│   ├── ...
│
├── ...

Usage

Exploring the Dataset: Researchers and enthusiasts can explore the dataset using various visualization tools to gain insights into African traditional attire.
Training Models: Data scientists and machine learning researchers can use this dataset to train models for various tasks like image classification, object detection, and style recognition.
Creative Projects: Artists, designers, and content creators can use these images as references for creating artwork, fashion designs, and more.

Citation

If you use this dataset in your research or projects, please consider citing it:

@dataset{african-attire-dataset,
  author = {inuwamobarak},
  title = {African Attire Images Dataset},
  year = {2023},
  publisher = {Inuwa Mobarak Abraham},
  howpublished = {\url{https://huggingface.co/datasets/inuwamobarak/african-atire}},
}

Contact

For inquiries, suggestions, or issues related to the dataset, please contact inuwamobarak@gmail.com.

Note: Respect copyright and licensing agreements when using the images from this dataset.