---
language:
- en
license: cc-by-nc-nd-4.0
task_categories:
- image-classification
- image-to-image
- object-detection
tags:
- medical
- biology
- code
dataset_info:
  features:
  - name: id
    dtype: uint16
  - name: front
    dtype: image
  - name: left_side
    dtype: image
  - name: right_side
    dtype: image
  - name: type
    dtype: string
  splits:
  - name: train
    num_bytes: 691862223.0
    num_examples: 30
  download_size: 691900290
  dataset_size: 691862223.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

# Skin Defects Dataset

The dataset contains images of individuals with various skin conditions: **acne, skin redness, and bags under the eyes**. Each person is represented by **3 images** showcasing their specific skin issue. The dataset encompasses diverse *demographics, age, ethnicities, and genders.*

**Types of defects in the dataset**: acne, skin redness & bags under the eyes

![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12421376%2Fe0f981eeb53fa147dcde53de5a91de7c%2FFrame%2056.png?generation=1700159939108017&alt=media)

- **Acne photos**: display different severities and types of acne such as whiteheads, blackheads, and cystic acne. 
- **Skin redness photos**: display individuals with this condition, which may be caused by rosacea or eczema. 
- **Bags under the eyes photos**: depicts individuals with noticeable bags under their eyes, often associated with lack of sleep, aging, or genetics.

## Full version of the dataset includes much more photos of people, leave a request on **[TrainingData](https://trainingdata.pro/datasets/dermatology?utm_source=huggingface&utm_medium=cpc&utm_campaign=skin-defects)** to buy the dataset

The dataset is a valuable resource for researchers, developers, and organizations working at the **dermatology, cosmetics and medical sphere** to train, evaluate, and fine-tune **AI models** for real-world applications. It can be applied in various domains like *skincare, scientific research and advertising*. 

# Get the Dataset
## This is just an example of the data 
Leave a request on **[https://trainingdata.pro/datasets](https://trainingdata.pro/datasets/dermatology?utm_source=huggingface&utm_medium=cpc&utm_campaign=skin-defects) to learn about the price and buy the dataset**

# Content
The folder **files** includes:
- **3 folders** with images of people with the conditions mentioned in the name of the folder (**acne, skin redness or bags under the eyes**)
- each folder includes sub-folders with **3 images** of each person from different angles: **front, left side and right side**

![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12421376%2F0c57df2bbfb6745dbfd5807984efb869%2FFrame%2055.png?generation=1700159929463259&alt=media)

### File with the extension .csv
- **id**: id of the person,
- **front**: link to access the front photo, 
- **left_side**: link to access the left side's photo,
- **right_side**: link to access the right side's photo,
- **type**: type of the defect (**acne, skin redness or bags under the eyes**)

## **[TrainingData](https://trainingdata.pro/datasets/dermatology?utm_source=huggingface&utm_medium=cpc&utm_campaign=skin-defects)** provides high-quality data annotation tailored to your needs

More datasets in TrainingData's Kaggle account: **https://www.kaggle.com/trainingdatapro/datasets**

TrainingData's GitHub: **https://github.com/Trainingdata-datamarket/TrainingData_All_datasets**

*keywords: biometric dataset, face recognition database, face recognition dataset, face detection dataset, facial analysis, dermatology dataset, skin on the face, IGA scale, medical data, whiteheads, blackheads, cystic acne, rosacea, eczema disease dataset, cosmetology, multi-task learning approach, facial acne image dataset, bumps on face, facial skin lesions, skin conditions, skin images, skin characteristics, automatic facial skin defect detection system, human face images, acne marks, stains, skincare, skin problems, skin disease dataset, human images, deep learning, computer vision*