tensorflow下的object detection從0開始訓練

主要參考：Custom Object Detection using TensorFlow from Scratch

安裝

• 安裝tensorflow

pip install tensorflow

• 克隆官方的object detection 倉庫

git clone https://github.com/tensorflow/models.git

• 設置環境

cd <path_to_your_tensorflow_installation>/models/research/
protoc object_detection/protos/*.proto --python_out=.
export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim

• 測試環境是否配置正確

python object_detection/builders/model_builder_test.py

• 下面顯示成功配置環境

[ RUN      ] ModelBuilderTest.test_create_rfcn_model_from_config
[       OK ] ModelBuilderTest.test_create_rfcn_model_from_config
[ RUN      ] ModelBuilderTest.test_create_ssd_fpn_model_from_config
[       OK ] ModelBuilderTest.test_create_ssd_fpn_model_from_config
[ RUN      ] ModelBuilderTest.test_create_ssd_models_from_config
[       OK ] ModelBuilderTest.test_create_ssd_models_from_config
[ RUN      ] ModelBuilderTest.test_invalid_faster_rcnn_batchnorm_update
[       OK ] ModelBuilderTest.test_invalid_faster_rcnn_batchnorm_update
[ RUN      ] ModelBuilderTest.test_invalid_first_stage_nms_iou_threshold
[       OK ] ModelBuilderTest.test_invalid_first_stage_nms_iou_threshold
[ RUN      ] ModelBuilderTest.test_invalid_model_config_proto
[       OK ] ModelBuilderTest.test_invalid_model_config_proto
[ RUN      ] ModelBuilderTest.test_invalid_second_stage_batch_size
[       OK ] ModelBuilderTest.test_invalid_second_stage_batch_size
[ RUN      ] ModelBuilderTest.test_session
[  SKIPPED ] ModelBuilderTest.test_session
[ RUN      ] ModelBuilderTest.test_unknown_faster_rcnn_feature_extractor
[       OK ] ModelBuilderTest.test_unknown_faster_rcnn_feature_extractor
[ RUN      ] ModelBuilderTest.test_unknown_meta_architecture
[       OK ] ModelBuilderTest.test_unknown_meta_architecture
[ RUN      ] ModelBuilderTest.test_unknown_ssd_feature_extractor
[       OK ] ModelBuilderTest.test_unknown_ssd_feature_extractor
----------------------------------------------------------------------
Ran 16 tests in 0.230s

OK (skipped=1)

準備階段

• 克隆代碼到官方model文件夾下

cd {Path_to_Model}
git clone https://github.com/wangrui1996/Tensorflow_Object_Detection_Examples example

• 進入examples目錄下，創建xmls目錄，並將標準後的文件放入其中

cd annotatinos 
mkdir xmls
# 進入example主目錄 創建images目錄，並將圖片放入此目錄中
mkdir images

分別執行tools目錄下的腳本

python3 create_tf_record.py

• 執行train.sh開始訓練

sh train.sh