CoDa

该数据集名为CoDa，包含了521个对象的查询以及人类对这些对象的颜色判断，这些颜色涵盖了11种基本英语颜色。数据集分为三种问题类型：单一、多重和任意，并在这些类别上报告了微平均结果。此外，该数据集还根据训练分类器的需要，进行了训练集、验证集和测试集的划分。规模上，它包含了521个对象的颜色分布。该数据集的任务是评估基于人类判断的物体典型颜色。

The dataset is named CoDa. It includes queries for 521 objects along with human color judgments for these objects, where the covered colors fall into 11 basic English color terms. The dataset is divided into three question types: single, multiple, and arbitrary, with micro-averaged results reported for each category. Additionally, the dataset has been split into training, validation, and test sets to support classifier training. In terms of scale, it contains the color distribution data of the 521 objects. The task of this dataset is to evaluate the typical colors of objects based on human judgments.