Human Face Emotions is an image classification dataset for the task of facial expression recognition, designed to provide basic training and evaluation data for research related to computer vision and emotion recognition.

This dataset is organized by emotion category, containing five facial emotion categories: Angry, Fear, Happy, Sad, and Surprised. Each category contains approximately 8,000+ facial image samples. The dataset uses a single facial image as the basic sample unit and is organized into folders based on emotion category. The image data comes from multiple publicly available online platforms, including Pexels, Google Images, Bing Images, Unsplash, and Kaggle, covering different people, shooting conditions, and background scenes, demonstrating good diversity.