deeplab V3 Mask RCNN tensorflow实现 语义分割 人物背景分开 合成不同场景的图像 目标检测

 

 

实现效果

 

mask RCNN  目标检测和识别

 

 

 

Deeplab V3 语义分割

提取检测目标

 

 

人物检测和提取

 

 

 

 

 

Mask RCNN

人物检测和显示

 

 

 

deeplab V3 人物和背景提取， 合成显示

 

 

 

 

 

 

 

 

 

代码整理好，再分享。

 

 

 

 

DeeplabV3

 

https://github.com/tensorflow/models/tree/master/research/deeplab

 

 

安装

 

替换deeplab 最新

 

 

######################

查看所有环境

conda info --envs

1

新建虚拟环境

conda create -n myenv python=3.6

1

删除虚拟环境

conda remove -n myenv --all

1

**虚拟环境

conda activate myenv

1

退出虚拟环境

conda deactivate myenv

 

创建环境

conda create -n TF_Deeplab --clone tensorflow-API

 

 

 

 

 

 

Pytest  安装

 

Numpy 更新1.15.1

 

 

 

 

运行model_test 测试

 

 

Quick test by running model_test.py:

# From tensorflow/models/research/

python deeplab/model_test.py

Quick running the whole code on the PASCAL VOC 2012 dataset:

# From tensorflow/models/research/deeplab

sh local_test.sh

DeepLab depends on the following libraries:

• Numpy
• Pillow 1.0
• tf Slim (which is included in the "tensorflow/models/research/" checkout)
• Jupyter notebook
• Matplotlib
• Tensorflow

 

We provide deeplab models pretrained several datasets, including (1) PASCAL VOC 2012, (2) Cityscapes, and (3) ADE20K for reproducing our results

Pretrained model

 

PASCAL2012

 

 

 

 

 

 

 

CITYSCAPES

 

 

ADE

 

 

 

Mobilenet

https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet