The Oxford-IIIT Pet Dataset

Oxford-IIIT宠物数据集是一个37类宠物数据集，每个数据类大约有100张图像，由牛津大学的Visual Geometry Group创建。图像在比例，姿势和照明方面有很大的差异。所有图像均具有相关的品种，头部ROI和像素级三图分割的地面真相注释。

The Oxford-IIIT Pet Dataset is a 37-category pet dataset, with approximately 100 images per category, created by the Visual Geometry Group at the University of Oxford. The images exhibit substantial variations in scale, pose and illumination. All images are accompanied by ground-truth annotations including breed labels, head ROI (Region of Interest) and pixel-level three-map segmentation.

搜集汇总

背景与挑战

背景概述

The Oxford-IIIT Pet Dataset是一个包含37类宠物、每类约100张图像的数据集，适用于语义分割、对象定位和对象检测任务。数据集由牛津大学Visual Geometry Group创建，图像在比例、姿势和照明方面有较大差异，且每张图像都有详细的注释信息。