前言
1. Labelme的安装
1.1 windows;
1.2 ubuntu;
conda create -n labelme python=3.8
conda activate labelme
pip install labelme
conda remove -n labelme --all
conda create -n labelme python=3.9
conda activate labelme
pip install
安装系统是ubuntu20.04,系统python版本是3.9,开始安装python3.8出现以下错误,之后删除环境重新安装python3.9,正常。
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
uff 0.6.9 requires protobuf>=3.3.0, which is not installed.
onnxruntime 1.11.1 requires protobuf, which is not installed.
onnx 1.11.0 requires protobuf>=3.12.2, which is not installed.
onnx-simplifier 0.3.10 requires protobuf>=3.7.0, which is not
注意,python2与python3的安装有些微区别;
2. LabelMe的使用;
2.1 windows;
启动:
标注:
文件格式转换:
2.2 ubuntu;
启动:打开labelme conda虚拟环境,输入labelme即可出现标注界面。
标注:
文件格式转换:
主要point:
1. 工具的安装,注意版本,不同的版本可能会出现问题;可以解决的;
2. 制作数据集的路径;
3. 最繁琐的就是标注;其次是生成数据集的过程;
参考:
1. 语义分割中单类别和多类别图片数据标注,以及灰度类别转换;
2. 数据注释-工具安装和使用:labelimg/labelme,标注;
完