SOLO: Segmenting Objects by Locations（SOLO and SOLO V2）

用法

A quick demo

安装完成后，下载训练好的models（在英文版readme.md），然后运行 inference_demo.py to run a quick demo.

多显卡

#  multiple GPUs trainng
./tools/dist_train.sh ${CONFIG_FILE} ${GPU_NUM}

Example: 
./tools/dist_train.sh configs/solo/solo_r50_fpn_8gpu_1x.py  8

# multi-gpu testing
./tools/dist_test.sh ${CONFIG_FILE} ${CHECKPOINT_FILE} ${GPU_NUM}  --show --out  ${OUTPUT_FILE} --eval segm

Example: 
./tools/dist_test.sh configs/solo/solo_r50_fpn_8gpu_1x.py SOLO_R50_1x.pth  8  --show --out results_solo.pkl --eval segm

单显卡

#  single GPU training
python tools/train.py ${CONFIG_FILE}

Example:
python tools/train.py configs/solo/solo_r50_fpn_8gpu_1x.py

## single-gpu testing
python tools/test_ins.py ${CONFIG_FILE} ${CHECKPOINT_FILE} --show --out  ${OUTPUT_FILE} --eval segm

Example: 
python tools/test_ins.py configs/solo/solo_r50_fpn_8gpu_1x.py  SOLO_R50_1x.pth --show --out  results_solo.pkl --eval segm

Visualization

python tools/test_ins_vis.py ${CONFIG_FILE} ${CHECKPOINT_FILE} --show --save_dir  ${SAVE_DIR}

Example: 
python tools/test_ins_vis.py configs/solo/solo_r50_fpn_8gpu_1x.py  SOLO_R50_1x.pth --show --save_dir  work_dirs/vis_solo

安装

安装以下包

Linux (只支持linux)
Python 3.5+
PyTorch 1.1 or higher
CUDA 9.0 or higher
NCCL 2
GCC 4.9 or higher
mmcv 0.2.16

我用的：

Python 3.6
PyTorch 1.4
CUDA 10.1
mmcv 0.2.16

具体操作：

安装新环境

conda create -n solo python=3.6
conda activate solo

安装cudatoolkit和cudnn

conda install cudatoolkit=10.1 cudnn

安装PyTorch 和 torchvision

conda install pytorch=1.4 torchvision

mmcv 0.2.16

pip install mmcv==0.2.16

安装SOLO包

git clone https://github.com/WXinlong/SOLO.git
cd SOLO
pip install -r requirements/build.txt
# 安装pycocotools（如果装不上就下下来，cd到PythonAPI文件夹，pip install setup.py）
pip install "git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI"
# 安装SOLO，倒腾完了
pip install -v -e .  # or "python setup.py develop"

SOLO以dev模式安装，对代码所做的任何本地修改都将生效，无需重新安装（除非您提交了一些提交并希望更新版本号）。

准备数据集

建议将数据集根目录符号链接到“$SOLO/data”。如果文件夹结构不同，需要更改配置文件中相应的路径。

SOLO
├── mmdet
├── tools
├── configs
├── data
│   ├── coco
│   │   ├── annotations
│   │   ├── train2017
│   │   ├── val2017
│   │   ├── test2017
│   ├── cityscapes
│   │   ├── annotations
│   │   ├── train
│   │   ├── val
│   ├── VOCdevkit
│   │   ├── VOC2007
│   │   ├── VOC2012

License

For academic use, this project is licensed under the 2-clause BSD License - see the LICENSE file for details. For commercial use, please contact Xinlong Wang and Chunhua Shen.

mepleleo / SOLO