[Software] environment to build
Operating System: windows 10 64 Wei
Memory: 8G
CPU:I7-6700
Tensorflow: 1.4
Python:3.5
Anaconda3 (64-bit)
Above environment to build your own to Baidu to view the tutorial.
[API] environment to build
1, download Tensorflow object detection API
Download: https://github.com/tensorflow/models
Because sometimes Github download fails, small API specially uploaded to Baidu cloud disk for you to download: https://pan.baidu.com/s/1cjgWwCGGoj30qjwS6Gbtyw Password: du86
2, Protobuf Installation and Configuration
Download: https://github.com/google/protobuf/releases (Please download window version: see below)
3, extract Protobuf, the bin folder protoc.exe into the C: \ Windows
4, the models-master at Tensorflow object detection API data / research \ directory window open command, enter:. Protoc object_detection / protos / * proto --python_out =.
Here will be prompted to find dialed the wrong file, you can try to /*.proto this part into a specific folder under the file name, one by one test, each run one in the corresponding folder should appear at the end of the corresponding .py document.
5, PYTHONPATH environment variable
In 'this computer' - 'Properties' - 'Advanced System Settings' - 'environment variable' - 'System Variables' of the new variable called 'the PYTHONPATH' of the Master-Models / Research / and models-master / research / slim two full directory folder to add, separated by semicolons, the following picture:
6, the test API
Then you can test API, run the command line in the models-master / research / Folder: not being given the instructions to run successfully.
python object_detection/builders/model_builder_test.py
[Case] test comes
1, open jupyter notebook
"Start -Anaconda3-Anaconda Prompt" to bring up the command line, change the working directory to models-master \ research \ object_detection
Note: There may be some people set jupyter notebook file path, you need to first comment out the default path. Specific methods, see: http://www.cnblogs.com/raorao1994/p/8058924.html
2, run the test code
Invoke the browser (Chrome) open the current folder, opening the object_detection_tutorial.ipynb, Object Detection Demo opens in a new tab, click on the top of the "Cell" - "Run All".
3. Results
Source acquisition mode, the total number of public attention RaoRao1994, wonderful view to the stage - all articles, we can get the download link resources
Get more resources, please pay attention to the public the total number RaoRao1994