3 labeled skin tone categories including White, Brown, and Black classes for image classification. The dataset provides a structured framework for training models to distinguish between specific human skin pigmentation levels.
Use Cases
- Train a deep learning model to categorize images into 'White', 'Brown', or 'Black' classes
- Analyze model fairness by measuring error rates across the three skin tone labels
- Automate the sorting of portrait photography based on the 'White', 'Brown', and 'Black' classification categories
Strengths
- Contains 3 distinct classification labels: White, Brown, and Black
- Organized for supervised learning tasks in the domain of skin tone analysis
- Provides categorical ground truth for human-centric computer vision datasets