[객체추출] Object detection with 텐서플로
Object detection using Tensor flow API
Setup (MAC)
Install Python library
$ pip install pillow
$ pip install lxml
$ pip install jupyter
$ pip install matplotlib
Install protocol buffer
1. download protobuf-all-21.5.tar.gz from https://github.com/protocolbuffers/protobuf/releases
2. Extract the tar.gz file.
3. $cd ~/Downloads/protobuf-2.4.1
4. $./configure
5. $make
6. $make check
7. $sudo make install
8. $which protoc
9. $protoc --version
Install Tensorflow object detection API
git clone https://github.com/tensorflow/models
Protocol buffer compile
1. go to the directory where API is installed
$cd models\research\object_detection
2. compile
$ protoc object_detection/protos/*.proto --python_out=.
3. configure path
$vi ~/.bash_rc
export PYTHONPATH=$PYTHONPATH:/PATH-TO/models/research:/PATH-TO/models/research/slim
$ source ~/.bash_rc
4. test
$python model_builder_test.py
if you see some errors, try :
$sudo pip3 install absl-py
$sudo pip3 install google
$sudo pip3 install protobuf
$sudo pip3 install six
$brew install tensorflow
$pip3 install tensorflow-object-detection-api --user
$xcode-select --install
Tutorial (Demo)
models/research/object_detection/object_detection_tutorial.ipynb
Application
1) add your images
add your test image in the ~/research/object_detection/test_images with the format of image{number}.jpg
change the TEST_IMAGE_PATHS range in Detectuon code block
2) use different pretrained models
download the model in github repository :
unzip the model and locate it to ~/models/research/object_detection/{model_folder_name}
/frozen_inference_graph.pb
change the MODEL_NAME in Variables block
3) create own labeling in the image
LabelImg
https://github.com/heartexlabs/labelImg
list of labeled image open dataset :
- voc2012
- coco
- imageNet
- Open Image Dataset V4