linkyle910125/FER2025

FER2025是一个大规模、平衡的面部表情识别数据集，专为深度学习和计算机视觉研究设计。数据集包含1,589,810张图片，覆盖7种情绪类别（愤怒、厌恶、恐惧、快乐、中性、悲伤、惊讶），每类图片数量差异不超过1.3%。数据集以TAR文件格式存储，每个TAR文件包含图片和对应的.cls标签文件，适合高效流式处理。推荐用途包括特征提取、模型训练和评估，以及大规模训练。数据集采用CC BY-NC 4.0许可，仅限非商业研究使用。

FER2025 is a large-scale, balanced facial emotion dataset designed for deep learning and computer vision research. It contains 1,589,810 images across 7 emotion classes (Angry, Disgust, Fear, Happy, Neutral, Sad, Surprise) with maximum class difference ≈ 1.3%. The dataset is stored in TAR files, each containing images + corresponding .cls label files for efficient streaming. Recommended uses include feature extraction, model training & evaluation, and large-scale training. The dataset is licensed under CC BY-NC 4.0 for non-commercial research use only.