FoodX-251 食物图像分类数据集

FoodX-251 是包含 251 个细粒度类别的食物图像数据集，有 120,216 张训练图像， 12,170 张验证图像和 228,399 张测试图像。训练图像带有杂音标签，验证图像和测试图像带有经人工验证的标签。这些种类是细粒度的且在视觉上是相似的，例如不同类型的蛋糕、三明治、布丁、汤和面食等。该数据集可用于研究细粒度可视化分类领域。

FoodX-251 is a fine-grained food image dataset encompassing 251 distinct categories. It comprises 120,216 training images, 12,170 validation images, and 228,399 test images. The training images are associated with noisy labels, whereas the validation and test images carry manually verified labels. These categories are fine-grained and visually similar, including various types of cakes, sandwiches, puddings, soups, pastas, and other similar food items. This dataset can be utilized for research in the domain of fine-grained visual classification.

搜集汇总

背景与挑战

背景概述

FoodX-251是一个包含251个细粒度类别的食物图像数据集，涵盖共计360,785张图像，其中训练图像带有杂音标签，验证和测试图像则经过人工验证。该数据集适用于细粒度视觉分类领域的研究，专注于视觉上相似的食物类别如蛋糕、三明治等。