mapiliary_vista
收藏官方服务：
资源简介：
包含了18k的训练集，2k的验证集和5k的测试集。一共有66类，可用于语义分割paddleseg的直接应用，已写好训练，验证，测试的txt文本。
This dataset comprises an 18k training set, a 2k validation set, and a 5k test set. It contains a total of 66 categories, and can be directly applied for semantic segmentation tasks using PaddleSeg. Pre-written training, validation, and test TXT files are already prepared.
提供机构：
hello_world007
