数据堂—10,100张人脸图像描述数据

10,100张人脸图像描述数据，包含年龄在18岁以下，18~45岁，46~60岁，60岁以上，的多人种；采集场景丰富，包含室内场景与室外场景；图像内容丰富，含有戴口罩、戴眼镜、戴耳机、表情、姿态以及对抗样本。文本描述语言为英语，主要描述了人种、性别、年龄、拍摄角度、光照以及多样性内容等

A dataset containing 10,100 facial image descriptions, covering multiple ethnic groups across four age brackets: under 18 years old, 18–45 years old, 46–60 years old, and over 60 years old. The dataset is collected from diverse scenarios including both indoor and outdoor environments. The images have rich content diversity, featuring subjects wearing face masks, eyeglasses, headphones, with various expressions and postures, as well as including adversarial examples. The accompanying textual descriptions are in English, mainly describing attributes such as ethnicity, gender, age, shooting angles, lighting conditions, and diversity-related contents.

搜集汇总

背景与挑战

背景概述

该数据集包含10,100张人脸图像及其描述，覆盖多种族、年龄、性别和场景，内容涵盖佩戴口罩、眼镜、表情等多方面。数据以.jpg图像和.txt文本格式提供，描述语言为英文和中文，标注准确率不低于97%。