美国手语中字母的图像数据集

数据集是来自美国手语的字母图像的集合，分为29个文件夹，代表各个类别。 训练数据集包含87,000张200x200像素的图像。有29个类，其中26个是字母AZ，3个是SPACE，DELETE和NOTHING。这3个类在实时应用程序和分类中非常有帮助。测试数据集仅包含29张图像，以鼓励使用真实世界的测试图像。

This dataset is a collection of alphabet images derived from American Sign Language (ASL), which is split into 29 folders corresponding to each category. The training dataset includes 87,000 images with a resolution of 200×200 pixels. There are 29 total categories, where 26 represent the letters A through Z, and the remaining three are SPACE, DELETE, and NOTHING. These three supplementary categories are highly valuable for real-world applications and classification tasks. The test dataset contains merely 29 images, intended to encourage the use of real-world test images.

搜集汇总

背景与挑战

背景概述

该数据集包含87,000张美国手语字母的200x200像素图像，分为29个类别（26个字母和3个特殊类别）。测试集包含29张图像，旨在鼓励使用真实世界测试图像。