Home

Awesome

<a href="https://doi.org/10.5281/zenodo.1248776"><img src="https://zenodo.org/badge/DOI/10.5281/zenodo.1248776.svg" alt="DOI"></a>

DeepLab_V3 Image Semantic Segmentation Network

Implementation of the Semantic Segmentation DeepLab_V3 CNN as described at Rethinking Atrous Convolution for Semantic Image Segmentation.

For a complete documentation of this implementation, check out the blog post.

Dependencies

Downloads

Evaluation

Pre-trained model.

Place the checkpoints folder inside ./tboard_logs. If the folder does not exist, create it.

Retraining

Original datasets used for training.

Place the tfrecords files inside ./dataset/tfrecords. Create the folder if it does not exist.

Training and Eval

Once you have the training and validation TfRefords files, just run the command bellow. Before running Deeplab_v3, the code will look for the proper ResNets checkpoints inside ./resnet/checkpoints, if the folder does not exist, it will first be downloaded.

python train.py --starting_learning_rate=0.00001 --batch_norm_decay=0.997 --crop_size=513 --gpu_id=0 --resnet_model=resnet_v2_50

Check out the train.py file for more input argument options. Each run produces a folder inside the tboard_logs directory (create it if not there).

To evaluate the model, run the test.py file passing to it the model_id parameter (the name of the folder created inside tboard_logs during training).

Note: Make sure the test.tfrecords is downloaded and placed inside ./dataset/tfrecords.

python test.py --model_id=16645

Retraining

To use a different dataset, you just need to modify the CreateTfRecord.ipynb notebook inside the dataset/ folder, to suit your needs.

Also, be aware that originally Deeplab_v3 performs random crops of size 513x513 on the input images. This crop_size parameter can be configured by changing the crop_size hyper-parameter in train.py.

Datasets

To create the dataset, first make sure you have the Pascal VOC 2012 and/or the Semantic Boundaries Dataset and Benchmark datasets downloaded.

Note: You do not need both datasets.

After, head to dataset/ and run the CreateTfRecord.ipynb notebook.

The custom_train.txt file contains the name of the images selected for training. This file is designed to use the Pascal VOC 2012 set as a TESTING set. Therefore, it doesn't contain any images from the VOC 2012 val dataset. For more info, see the Training section of Deeplab Image Semantic Segmentation Network.

Obs. You can skip that part and direct download the datasets used in this experiment - See the Downloads section

Serving

For full documentation on serving this Semantic Segmentation CNN, refer to How to deploy TensorFlow models to production using TF Serving.

All the serving scripts are placed inside: ./serving/.

To export the model and to perform client requests do the following:

  1. Create a python3 virtual environment and install the dependencies from the serving_requirements.txt file;

  2. Using the python3 env, run deeplab_saved_model.py. The exported model should reside into ./serving/model/;

  3. Create a python2 virtual environment and install the dependencies from the client_requirements.txt file;

  4. From the python2 env, run the deeplab_client.ipynb notebook;

Results

Results