A Train Test demo of maskrcnn_opencv
- python3
- opencv (>3.4.3)
- tensorflow_gpu (12.0)
- cuda9.0 cudnn7.0
- install labelme
sudo pip3 install labelme
- generate tf.record
python3 create_tf_record.py \
--images_dir=path_to_images_dir \
--annotations_json_dir=path_to_train_annotations_json_dir \
--label_map_path=path_to_label_map.pbtxt \
--output_path=path_to_train.record
git clone https://github.com/tensorflow/models.git
在models/research/object_detection目录下执行
python model_main.py \
--model_dir=path/to/save/directory \
--pipeline_config_path=path/to/mask_rcnn_inception_v2_xxx.config
get model graph txt
python tf_text_graph_rcnn.py