Kvasir Landmark and Pathological Classification Dataset

该数据集包含多个解剖学标志和病理发现，共分为七个类别，包括标志点检测和病理发现，每个类别都有总共1000张独特的图像。在数据处理上，70%的数据用于训练，10%用于验证，20%用于测试。此外，每个帧的深度是使用基于深度学习的方法估算的。规模上，每个类别有1000张独特图像，总计7000张图像。该任务的目的是进行解剖学标志和病理分类。

This dataset includes multiple anatomical landmarks and pathological findings, and is divided into seven categories covering both landmark detection and pathological finding-related tasks. Each category contains 1,000 unique images, resulting in a total of 7,000 images overall. For dataset splitting, 70% of the data is allocated to the training set, 10% to the validation set, and 20% to the test set. Furthermore, the depth of each image frame is estimated via a deep learning-based approach. The objective of this task is to conduct anatomical landmark and pathological classification.