Skip to content

Tensorflow Faster R-CNN for Windows/Linux and Python 3 (3.5/3.6/3.7)

License

Notifications You must be signed in to change notification settings

whuyyc/Faster-RCNN-TensorFlow-Python3

 
 

Repository files navigation

tf-faster-rcnn

Tensorflow Faster R-CNN for Windows and Linux by using Python 3

This is the branch to compile Faster R-CNN on Windows and Linux. It is heavily inspired by the great work done here and here. I have not implemented anything new but I fixed the implementations for Windows, Linux and Python 3.

Currently, this repository supports Python 3.5, 3.6 and 3.7. Thanks to @morpheusthewhite

PLEASE BE AWARE: I do not have time or intention to fix all the issues for this branch as I do not use it commercially. I created this branch just for fun. If you want to make any commitment, it is more than welcome. Tensorflow has already released an object detection api. Please refer to it. https://github.com/tensorflow/models/tree/master/research/object_detection

If you find a solution to an existing issue in the code, please send a PR for it.

Also, instead of trying to deal with Tensorflow, use Chainer. It is ready to be used with all the common models https://github.com/chainer/chainercv & https://github.com/chainer/chainer . I can reply all of your questions about Chainer

How To Use This Branch

  1. Install tensorflow, preferably GPU version. Follow instructions. If you do not install GPU version, you need to comment out all the GPU calls inside code and replace them with relavent CPU ones.

  2. Checkout this branch

  3. Install python packages (cython, python-opencv, easydict) by running
    pip install -r requirements.txt
    (if you are using an environment manager system such as conda you should follow its instruction)

  4. Go to ./data/coco/PythonAPI
    Run python setup.py build_ext --inplace
    Run python setup.py build_ext install
    Go to ./lib/utils and run python setup.py build_ext --inplace

  5. Follow these instructions to download PyCoco database. I will be glad if you can contribute with a batch script to automatically download and fetch. The final structure has to look like
    data\VOCDevkit2007\VOC2007

  6. Download pre-trained VGG16 from here and place it as data\imagenet_weights\vgg16.ckpt.
    For rest of the models, please check here

  7. Run train.py

Notify me if there is any issue found.

About

Tensorflow Faster R-CNN for Windows/Linux and Python 3 (3.5/3.6/3.7)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 96.0%
  • MATLAB 2.0%
  • Lua 0.7%
  • C++ 0.7%
  • C 0.5%
  • Jupyter Notebook 0.1%